CELLULITIS IN TURKEYS: CHARACTERIZATION OF …

CELLULITIS IN TURKEYS: CHARACTERIZATION OF CAUSATIVE

AGENTS AND PREVENTIVE MEASURES

A DISSERTATION

SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL

OF THE UNIVERSITY OF MINNESOTA

BY

ANIL JOHNY THACHIL

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

ADVISER: KAKAMBI V. NAGARAJA

AUGUST, 2011

© Anil J. Thachil, 2011

ACKNOWLEDGEMENTS

I would like to begin by acknowledging my mentor and advisor Dr. Kakambi V.

Nagaraja for his excellent mentorship. I cannot express in words his support,

encouragement, patience, and financial assistance through out the period of my study. I

have greatly benefitted from his breadth of experience and his teachings and valuable

suggestions will always help in my future scientific career.

I thank Dr. D.A. Halvorson for his meticulous involvement in my research and

helped me substantially in accomplishing this goal. I express my sincere thanks to Dr.

Sally Noll, Dr. Sagar Goyal and Dr. Timothy Johnson for their guidance, support and

constructive criticism and help extended as members of my thesis committee.

Also invaluable in the completion of this thesis was the contributions made by

turkey field veterinarians Drs Carl Heeder, Brian McComb, and Michelle Andersen. I

thankfully acknowledge the expertise and collaboration of Dr. Daniel Shaw at the

University of Missouri-Columbia in evaluating my histopathology slides. I thank Dr.

Srinand Sreevatsan and the members of his laboratory for a great deal of help in training

me and trusting me with their equipment. Thanks are due to Drs. Mathur Kannan, and

Samuel Maheswaran for their help and constant encouragement.

The help and support of the staff of the Department of Veterinary and Biomedical

Sciences, Research Animal Resources, and Minnesota Veterinary Diagnostic Laboratory

are highly appreciated.

I have been fortunate to have had the support of wonderful friends and colleagues

during these years. These include Drs Binu Velayudhan, Yogesh Chander, Devi

Patnayak, Antony Jude, and Dileepan Thamotharampillai who have been my colleagues

i

and close personal friends, without whom work would have been significantly difficult

and infinitely more dreary. Arpita Ghosh, Igor Radovic, Drs Vanessa Lopes and, Sunil

Kumar and have all been good friends who have contributed to my journey and to whom

I owe a debt of gratitude.

I would like to extend my thanks to Minnesota Turkey Research Promotion

Council, US Egg and Poultry Association, and Minnesota Agricultural Experiment

Station for their financial support.

I would also like to express my gratitude to my parents who helped me embark on

this opportunity and who have always believed in my ability to complete it. Special

thanks are due to my dear wife Toncy Edakalathoor and my daughter Anlin Thachil for

their unconditional love, forbearance and strength through this process.

ii

Dedicated

to

my parents

D.r T. K. Johny and Prof.Pushpi Johny

iii

ABSTRACT

Cellulitis continues to cause extensive losses in turkey production in USA due to

severe mortality, carcass condemnation and treatment costs. Clostridium perfringens and

Clostridium septicum have been recognized as the primary causative agents of cellulitis in

turkeys. In this study, cellulitis lesions and mortality in turkeys were successfully

reproduced with Clostridium perfringens and Clostridium septicum isolated from

cellulitis cases. Clostridium perfringens and Clostridium septicum isolates varied in their

ability to produce spores as well as toxins. We observed differences in the toxicity and

biological effects of different strains of C. perfringens and C. septicum in vitro, and in vivo.

.Though the spore count and hemolytic effects of C. perfringens were found to be higher

than C. septicum in vitro, mortality studies in mice and turkeys showed that C. septicum

was much more potent than C. perfringens. However, gross lesions produced by C.

perfringens and C. septicum were almost identical. Surprisingly, the development of

cellulitis lesions and mortality was markedly higher in 7-week-old birds than in 3-week-

old birds. The results of our study demonstrated for the first time that both C. perfringens

and C. septicum can multiply in the subcutaneous and muscle tissues and cause cellulitis

lesions in turkeys. Our cellulitis disease model offers promise as a challenge model in the

development of vaccines against cellulitis in turkeys. Both bivalent C. perfringens and C.

septicum toxoid and C. septicum toxoid were found to be safe and offered complete

protection against cellulitis following homologous challenge under experimental

conditions. The use of these vaccines enabled us to reduce the mortality and antibiotic

usage in preventing cellulitis in commercial turkeys. Multiple vaccinations or use of a day

old vaccine followed by a booster dose probably will offer better protection than a single

iv

vaccination at 6-weeks of age against cellulitis due to C. perfringens and C. septicum in

turkeys.

v

TABLE OF CONTENTS

ACKNOWLEDGEMENTS…………………………………..…………………….. i

DEDICATION…………………………………..…………………………………... iii

ABSTRACT…………………………………..……………………………………… iv

TABLE OF CONTENTS……………………..…………………………………….. vi

Chapter I. GENERAL INTRODUCTION………………………………………… 1

Chapter II. LITERATURE REVIEW …………………………………………….. 7

Cellulitis

Etiological agents attributed to Cellulitis

Clostridium perfringens

Clostridium septicum

Vaccination against Clostridial diseases

Chapter III. CHARACTERIZATION OF CLOSTRIDIUM PERFRINGENS AND

CLOSTRIDIUM SEPTICUM ISOLATES FROM CELLULITIS CASES.. 18

Chapter IV. ROLE OF CLOSTRIDIUM PERFRINGENS AND CLOSTRIDIUM

SEPTICUM IN CELLULITIS IN TURKEYS……………………………. 46

Chapter V. LABORATORY STUDIES ON A BIVALENT VACCINE

CONTAINING CLOSTRIDIUM PERFRINGENS AND CLOSTRIDIUM

SEPTICUM TOXOID TO CONTROL CELLULITIS IN TURKEYS…. 63

vi

vii

Chapter VI. FIELD TRIALS WITH BIVALENT CLOSTRIDIUM PERFRINGENS

AND CLOSTRIDIUM SEPTICUM TOXOID……………………............... 76

Chapter VII. LABORATORY STUDIES ON CLOSTRIDIUM SEPTICUM

TOXOID TO CONTROL CELLULITIS IN TURKEYS………………… 88

Chapter VIII. FIELD TRIALS WITH CLOSTRIDIUM SEPTICUM TOXOID.. 97

Chapter IX. CONCLUSIONS……………………………………………………... 110

Chapter X. REFERENCES………………………………………………………… 114

APPENDIX: AUTHORIZATION OF USE OF PUBLISHED MATERIAL…… 126

CHAPTER I

GENERAL INTRODUCTION

1

Over the last few years, cellulitis has become more prevalent in commercial

turkeys especially in Minnesota, Wisconsin, Missouri and Virginia. Accompanied by

high mortality and carcass condemnation in the processing plants, it often inflicts heavy

economic loss for the turkey producers (Susan, 2003). Because of this, cellulitis is

currently considered as a high priority research topic. The mortality is reported to be as

high as 1-2% per week in the affected flocks.

A turkey health survey (Clark et al., 2010) of US veterinarians in turkey

production ranked cellulitis at a score of 3.5; the survey ranked 34 current disease issues

(1= no issue to 5 = severe problem) and had a survey response (reply) of 100% (n=19).

During the 2007 Annual Convention of National Turkey Federation (NTF) in Tucson,

Arizona, the attendees recommended that NTF host a series of meetings to further discuss

turkey cellulitis. To help address this request, the Minnesota Turkey Growers Association

held a meeting during the Midwest Poultry Federation Convention to further discuss this

issue. The discussion focused on current research available and knowledge gaps and

research needs. The meeting concluded with the attendees making several

recommendations to the NTF Turkey cellulitis Task Force as it works toward better

understanding the disease and developing a control strategy. In the year 2008, a Gold

medal panel on cellulitis was held on December 16-17, Bloomington, MN, to invite

suggestions for control measures for cellulitis in turkey production. Development of an

effective vaccine against cellulitis was recognized as one of the urgent measures

necessary to control cellulitis in commercial turkeys. The panel also suggested renaming

the disease condition “Cellulitis” to “Clostridial dermatitis”.

2

Diagnostic laboratories have consistently isolated Clostridium perfringens and

Clostridium septicum organisms from turkey cellulitis lesions. The mode of entry of these

organisms is uncertain because in most cases of turkey cellulitis, there are no external

wounds that would explain the entry of the organism into the bird. It appears that entry of

the organisms into the bird is by oral route through contaminated feed.

It is widely accepted that natural outbreaks of cellulitis in turkeys are associated

with proliferation of pathogens in the poultry environment. One can argue that a reasonable

model of the disease should closely resemble the probable route of natural pathogen

exposure, i.e.: oral. Necrotic enteritis is yet another disease in chickens caused by C.

perfringens. Interestingly, efforts for consistent reproduction of necrotic enteritis by oral

inoculation of C. perfringens has resulted in extremely variable results including severe

clinical signs, subclinical NE and no lesions at all in exposed birds (Al-Sheikly et al., 1977;

Chalmers et al., 2007; Pederson et al., 2008; Truscott et al., 1977). Such a significant lack

of consistency in scientific data generated in different labs is puzzling indeed, particularly

in view of a large number of claims identifying predisposing factors (Mcdevitt et al., 2006;

Olkowski et al., 2006). Attempts to reproduce cellulitis in turkeys by injecting purified

alpha toxin of C. perfringens subcutaneously were successful but injections with C.

perfringens alone were not (Carr, 1996).

The incidence of Clostridium perfringens-associated diseases in poultry has

increased significantly in the recent years because of the reduced use of antimicrobial

growth promoters (van-Immerseel et al., 2004). It has become a common practice to use

antimicrobial drugs in feed or water to control cellulitis in a preventative manner, but this

practice is increasingly criticized or has been banned in some countries. Since cellulitis

3

cases in turkeys are also rising at an alarming rate, there is a need to investigate alternative

approaches for its effective control.

Vaccines containing toxoids and killed bacteria have been successfully used in

humans and animals against clostridial infections. Formalin-inactivated C. difficile toxoid

was found to be highly safe and immunogenic against C. difficile infections in mice (Torres

et al., 1995) and humans (Kotloff et al., 2001). For protection of birds affected with

necrotic enteritis caused by C. perfringens, alpha toxin attracted the most attention in

earlier studies but recent studies questioned the role of alpha toxins (Thompson et al.,

2006).

Despite the importance of cellulitis in turkeys, there is very little known about the

basis of immunity to this infection, although immunization is an obvious approach to

control. The purposes of this study were therefore to examine the aspects of immunity to

cellulitis in turkeys, specifically; to develop a challenge model to test any vaccines for

cellulitis in turkeys and to test whether it is possible to immunize turkeys against cellulitis.

The following three main objectives are there in this dissertation research.

I. To characterize isolates of Clostridium perfringens and Clostridium septicum from

cases of cellulitis in turkeys.

To achieve this objective the following experiments were carried out.

1. Laboratory characterizion of C. perfringens and C. septicum isolates using heat

resistant spore count, hemolytic assay and mice lethal assay.

4

2. Conducted Multilocus sequence typing (MLST) to determine the genetic

variation among C. perfringens and C. septicum isolated from cases of cellulitis in

turkeys

3. Identified secretory components of C. perfringens and C.septicum using 2-

DiGE analysis and MALDI-TOF mass spectrometry.

II. To study the role of Clostridium perfringens and Clostridium septicum in causing

cellulitis in turkeys.


1. To conduct in vivo studies to examine Clostridium perfringens and

Clostridium septicum in the cause of cellulitis in turkeys of different age groups.

III. To examine the use of Clostridium perfringens and/or Clostridium septicum

inactivated vaccines to prevent cellulitis in turkeys.


5

1. Prepared and examined an experimental bivalent Clostridium perfringens and

Clostridium septicum toxoid for prevention of cellulitis in turkeys under

experimental conditions.

2. Field tested the lab evaluated experimental Clostridium perfringens and

Clostridium septicum toxoid for its use in minimizing losses in turkey flocks in

Minnesota with existing cellulitis problem.

3. Prepared and examined an experimental Clostridium septicum toxoid for

prevention of cellulitis in turkeys under experimental conditions.

4. Field tested the lab evaluated experimental Clostridium septicum bacterin

toxoid for its use in minimizing losses in turkey flocks in Minnesota with existing

cellulitis problem.

6

CHAPTER II

LITERATURE REVIEW

7

Over the last several years cellulitis has emerged as an economically important

disease syndrome in turkeys. Accompanied by high mortality and carcass condemnation

in the processing plants, cellulitis often inflicts heavy economic loss for the turkey

producers (Olkowski et al., 1999; Susan., 2003). In recent years, cellulitis received more

attention because of an alarming upward trend in the mortality in turkey grower flocks

(Kumar et al., 1998). In a study conducted at a turkey processing plant in Ontario,

Canada, 9% of the birds examined had severe cellulitis (St. Hilaire et al., 2003).

Currently, cellulitis in turkeys is being diagnosed in Minnesota, Wisconsin, Missouri,

Virginia and other turkey producing areas. In Minnesota, the mortality is reported to be as

high as 1-2% per week in the affected flocks.

A. CELLULITIS

Cellulitis is described in turkeys as a condition characterized by inflammation of

the skin and subcutaneous tissue with accumulation of focal yellow or yellow-brown

exudate in subcutis of breast and tail areas (Carr et al., 1996; Kumar et al., 1998). Similar

disease condition has been described in chickens with a different name “Gangrenous

dermatitis” (van-Immerseel et al., 2004; Li et al., 2010b).

Only few references are available for clostridial infections in turkeys although the

first case was reported as early as in 1939 (Fenstermacher et al., 1939). Suspected cases of

Clostridial cellulitis have been observed in breeder toms following semen collection

(Ficken et al., 1991). In turkeys, Clostridial cellulitis and mortality have been reported

both in range and in breeder toms (Carr, 1996; Ficken et al., 1991) and also in hens (Carr et

al., 1996). It was determined to have been initiated by infection of wounds that the hens

8

received during natural mating. The lesions associated with cellulitis usually appear at 13

to 16 weeks of age and persist until the birds are marketed (Carr et al., 1996).

Cellulitis in turkeys is characterized by edematous swelling of breast, neck, tail

and legs, and accumulation of gaseous serosanguinous exudate within the subcutaneous

tissue (Fenstermacher et al., 1939; Carr et al., 1996). In addition, the lesions are seen also

on the back of the bird. Palpitation of the affected areas often reveals crepitation due to gas

bubbles in the subcutis and musculature. On necropsy, one would find accumulation of

bubbly, serosanguinous fluid in the subcutis. The affected tissues sometimes contain large

amounts of gelatinous exudate. The underlying musculature may have a cooked

appearance. The liver and spleen are often enlarged and may contain large necrotic infarcts.

The kidneys are usually swollen (Carr, 1996). Interestingly, in most cases of cellulitis,

there appears to be no obvious damage or trauma to the skin. Hence the role of any

pathogen isolated from these lesions in the causation of the disease remains obscure

(Olkowski et al., 1999). Not much information is available about the pathogenesis of

cellulitis in turkeys.

In chickens, cellulitis conditions caused by Clostridium septicum, and Clostridium

perfringens has been described previously as early as in 1965 (Saunders and Bickford,

1965). Severe outbreaks of cellulitis by C. septicum alone have also been reported in

broiler chickens (Helfer et al., 1969; Willooghby et al., 1996). In chickens with

gangrenous dermatitis, the wings, breast, abdomen and thighs are found to be commonly

affected (Li et al., 2010b). It is frequently found in birds of 4 to 16 weeks of age. The

disease is characterized by a sudden onset, an increase in mortality, gangrenous necrosis of

the skin over the thighs and breast, and an accumulation of serosanguinous fluid in the

9

subcutaneous tissue of the breast and thighs (Gerdon et al., 1973; Hofacre et al., 1986).

The skin in these areas appears blackened, moist, and devoid of feathers. There will be

bloody fluid and emphysema in the subcutis. The underlying musculature is often tan or

gray, emphysematous and may contain fluid between muscle bundles. Death occurs

within hours (Ficken, 1991; Hofacre et al., 1986). Many cases of gangrenous dermatitis

are believed to be a sequela to diseases that cause immunosuppression, such as infectious

bursal disease, infectious anemia or avian adenovirus infection (Ficken, 1991).

B. ETIOLOGICAL AGENTS ATTRIBUTED TO CELLULITIS

Infections with Clostridium perfringens, and Clostridium septicum is reported to

cause cellulitis in turkey breeder hens (Fenstermacher et al., 1939). Organisms isolated

from the cellulitis lesions of turkeys in Minnesota in the recent years also primarily include

C. perfringens type A and C. septicum either alone or in combination (personal

communication, Brian McComb, 2009). The isolation of either C. perfringens type A or C.

septicum or both from cases of cellulitis in chickens and turkeys has been emphasized in

many reports (Carr, 1996; Hofacre et al., 1986). Clostridium perfringens has been

cultured from gangrenous dermatitis lesions in broiler chickens (Saunders and Bickford,

1965) and is also identified as the organism responsible for the pathology associated with

necrotic enteritis in broilers (Hofacre et al., 1986; McDevitt et al., 2006) as well as

cellulitis in turkeys (Carr et al., 1996; Fenstermacher et al., 1939). Thus it is worth noting

that Clostridium perfringens is associated with distinct disease manifestations in birds but

the interesting observation here is that all three disease conditions appear distinct.

10

Occasionally E. coli were also isolated in low numbers from few cases of cellulitis in

turkeys (Gomis et al., 2002).

The etiology of gangrenous dermatitis in broiler chickens is attributed to multiple

bacterial pathogens. However, Escherichia coli are considered as one of the major

pathogens causing gangrenous dermatitis in broiler chickens (Gomis et al., 1997; Leclerc

et al., 2003; Messier et al., 1993). A unique subset of E. coli which is responsible for

cellulitis in broiler chickens was identified later (Jeffrey et al., 2002). Concurrent

infections with C. septicum and Staphylococcus aureus in cases with gangrenous

dermatitis in 4-week-old broiler chickens has also been reported (Wilder et al., 2001). In

addition, Frazier et al (1964) isolated E. coli, Streptococcus fecalis, Proteus sp., Bacillus

sp., Staphylococcus aureus and Clostridium septicum from cases of gangrenous dermatitis

in broiler chickens. Gangrenous dermatitis in broiler chickens has also been reported to be

associated with C. perfringens Type A and C. septicum or with C. novyi either alone or in

combination with C. perfringens, C. septicum, and Staphylococcus aureus (Susan, 2003).

Clostridium perfringens and C. septicum are both spore forming bacteria and their

spores are highly stable and ubiquitously present in the environment. These pathogens

are capable of producing a myriad of extracellular toxins and enzymes that degrade host

tissues and are responsible for the necrotic lesions observed (Sawires et al., 2006;

Smedley et al., 2004).

C. CLOSTRIDIUM PERFRINGENS

A group of anaerobic organisms affecting animals described as histotoxic

clostridia are most commonly involved in gas gangrene, or myonecrosis secondary to

11

wound infections. This group includes: C. perfringens, C. septicum, C. novyi, C.

histiolyticum, and C. bifermentans (Onderdonk et al., 1995). Clostridium perfringens is

one of the most widespread of all pathogenic bacteria affecting humans and animals. Its

main habitats are the soil and the intestinal tracts of animals and man. Armed with an

arsenal of many potent extra-cellular toxins and enzymes, C. perfringens is recognized as

the causative agent of human gas gangrene and food poisoning as well as several

enterotoxemic diseases in livestock (Johansson et al., 2006).

Clostridium perfringens is found in the intestinal tract of healthy poultry as a

normal inhabitant, usually in low numbers and has been isolated from processed carcasses

as well as from the processing plants (Craven et al., 2001b; Martel et al., 2004). Hatcheries

are identified as a potential source and reservoir for C. perfringens in integrated poultry

operations (Craven et al., 2001a). Experimental inoculation of purified alpha toxin of C.

perfringens produced lesions similar to cellulitis and also severe mortality in turkeys (Carr,

1996).

Necrotic enteritis and gangrenous dermatitis are among the most common

Clostridial diseases observed in broiler chickens which are caused by Clostridium

perfringens. In turkeys, clostridial cellulitis which is similar to gangrenous dermatitis in

chickens is the major disease caused by C. perfringens type A and C. septicum. It is

interesting to note that though both necrotic enteritis and gangrenous dermatitis in

broilers are attributed to C. perfringens type A, both disease conditions are not yet

reported in any flocks or in any farms at the same time (McDevitt et al., 2006). Moreover,

the incidence of one or the other disease condition continues to be more pronounced in

some well maintained farms than in others in spite of adopting strict control strategies. It

12

appears that the C. perfringens responsible for each disease conditions are different.

Clostridium perfringens isolates are grouped into types A through E based on

their production of one of the four major lethal toxins; alpha, beta, epsilon, and iota

(Sawires et al. 2006; Smedley et al., 2004). The growth of C. perfringens is restricted to

the site of infection, whereas the alpha toxin produced is absorbed by circulatory system

and causes massive intravascular hemolysis and destruction of capillary walls. All five

types of C. perfringens produces alpha-toxin which is an enzyme phospholipase C also

called as lecithinase (Karasawa et al., 2003; Smedley et al., 2004). In addition, C.

perfringens produces an array of extracellular toxins including beta2 toxin, enterotoxin,

perfringolysin, collagenase, lambda toxin, hyaluronidase, dnase, neuraminidase and

urease (Sawires et al. 2006). Alpha toxin is markedly hemolytic and it is the only major

toxin produced by C. perfringens type A (Titball et al., 1999). The sporulation process in

C. perfringens is reported to be linked with alpha-toxin production (Varga et al., 2004).

Clostridium perfringens produces more than twenty different types of toxins and

the toxin profile varies with the strain of bacteria (Meer et al., 1997; Shane et al., 2002).

Among toxins, alpha toxin was believed to be the predominant toxin responsible for C.

perfringens pathogenicity as well as protection. But recent studies indicate that Alpha

toxin does not have any role in the pathogenesis of necrotic enteritis caused by C.

perfringens nor do they protect birds as demonstrated with an alpha toxin mutant of C.

perfringens (Keyburn et al., 2006). Thompson et al., (2006) demonstrated that

immunogens other than alpha-toxin are important in protective immunity against C.

perfringens infection in broiler chickens.

13

Multi locus sequence typing (MLST) analysis of 132 isolates of C. perfringens

from different hosts identified 80 sequence types (ST) (Jost et al., 2006). Of all the loci

examined, plc gene had the most alleleles of 37 (Jost et al., 2006).

Recently many hypothetical proteases and β2-toxins were found in virulent C.

perfringens type A strains and that is believed to play a major role in eliciting a protective

immune response against necrotic enteritis. A novel toxin NetB is now identified as a

definitive virulence factor present in avian C. perfringens strains capable of causing

necrotic enteritis in chickens (Keyburn et al., 2008). Another group of scientists recently

demonstrated a hypothetical protein of 117 kDa suggestive of a protease found only in

virulent C. perfringens type A causing necrotic enteritis (Kulkarni et al., 2006).. This

protein is believed to play a major role in the development of necrotic enteritis and also in

eliciting a protective immune response (Kulkarni et al., 2007). It is also shown that the

β2-toxin gene (cpb2) found in isolates of C. perfringens cultured from avian hosts are

atypical compared to those found in pigs (Jost et al., 2005).

D. CLOSTRIDIUM SEPTICUM

Clostridium septicum is an organism of considerable medical and veterinary

importance. Little is known about the distribution and sources of C. septicum in poultry

production facilities. Clostridium septicum has played an important role as an etiologic

agent of traumatic gas gangrene and clostridium myonecrosis in humans (Smith-Slatas et

al., 2006). In mammals, particularly ruminants, a condition known as malignant edema is

caused by wound contamination with Clostridium septicum. Clostridium septicum also

appears to be a significant contributor to the etiology of cellulitis in turkeys. However,

14

not much information is available regarding the influence or the disease causing potential

of different strains of C. septicum.

Clostridium septicum produces four major toxins (alpha, beta, gamma and delta)

that are responsible for tissue damage and toxemia (Timoney et al., 1998). These toxins

include: the lethal necrotizing and hemolytic toxin (alpha-toxin); dnase (beta-toxin);

hyaluronidase (gamma-toxin); and the thiol-activated toxin or septicolysin (delta toxin)

(Cortinas et al., 1997). Other enzymes such as protease and neuraminidase are also

produced by C. septicum. Alpha-toxin plays a major role in C. septicum infection and is

also the major protective antigen. This toxin is produced as an inactive protoxin of

approximately 48 KD and requires proteolytic processing for activation into cytolytically

active protein. Hemolytic activities have been described for alpha-toxin and a thiol-

activated cytolysin produced by C. septicum.

The alpha-toxin of C. septicum is not related to the alpha toxin of C. perfringens

since it does not exhibit phospholipase C activity and does not cross-react

immunologically with C. perfringens alpha-toxin (Ballard et al., 1992). For in vitro toxin

production by C. septicum Brain heart infusion (BHI) medium is reported to be ideal

(Ballard et al., 1992).

E. VACCINATION AGAINST CLOSTRIDIAL DISEASES

Clostridial vaccines containing toxoids and killed bacteria have been successfully

used in humans and animals against clostridial infections. Toxoids against many

Clostridial diseases like Black Quarter (Clostridium chauvoei) and tetanus (C. tetani)

were found to be very safe and highly effective. Formalin-inactivated C. difficile toxoid

15

vaccine was found to be highly safe and immunogenic against C. difficile infections in

humans (Kotloff et al., 2001). Clostridium perfringens infections have been successfully

controlled in suckling piglets using a toxoid vaccine (Springer et al., 1999). Sheep

challenged with C. perfringens toxoid was found to be protective against gas gangrene

caused by C. perfringens (Boyd et al., 1972). A recombinant C. perfringens alpha toxoid

is also found to be protective against Clostridial myonecrosis in mice studies (Neeson et

al., 2007).

Given the key roles that some phospholipases play in the pathogenesis of many

diseases, it is not surprising that inactivated forms of these secretory toxins have been

considered as components of vaccines. The use of formaldehyde toxoids prepared from the

secretory toxins present in the supernatant of the culture of C. perfringens has induced

protection against necrotic enteritis in chickens. Immunity to necrotic enteritis in broiler

chickens was found to be associated with the presence of antibodies against C. perfringens

alpha toxin (Heier et al., 2001). In another study alpha toxin of C. perfringens

(phospholipase C) was found to be protective against lethal infections caused by C.

perfringens in mice (Stevens et al., 2004).

For protection of birds affected with necrotic enteritis caused by C. perfringens,

alpha toxin attracted the most attention in earlier studies but recent studies questioned the

role of alpha toxins (Thompson et al., 2006). Turkeys vaccinated with commercial C.

perfringens toxoid made for necrotic enteritis (sheep vaccine) did not protect against

cellulitis in turkeys (personal communication Brian McComb). It is conceivable that

there are genomic differences in C. perfringens strains that affect their phenotypic

expression of virulence and/or toxin production.

16

Genotyping methods will help us to distinguish bacterial strains. A multilocus

sequence typing (MLST) of C.perfringens isolates from porcine origin identified three

clonal complexes and eight sequence types by looking at the patterns of genetic

polymorphism in the house-keeping genes (colA, gyrA, plc, pfoS, and rplL) (Jost et al.

2006).

Vaccination has long been used to protect cattle and sheep against C. septicum

infection too. The difficulties in the preparation of C. septicum toxin cultures are well

known. Cultures often produce lethal antigen only in low titre and, in addition, the

immunogenicity of native toxin filtrates is weak, which results in poor antibody response

in animals (Cortinas et al., 1997). Formalin inactivated C. septicum alpha toxoid was

found to be protective against experimental challenge with C. septicum spores in mice

(Amimoto et al., 2002). A polyvalent C. perfringens, C. septicum and Pasteurella

bacterin was found to reduce mortality in gangrenous dermatitis affected broiler chickens

(Gerdon et al., 1973).

In this respect, the C. perfringens and C. septicum alpha toxins attracted the most

attention. However, recent studies (Kulkarni et al., 2006, Thompson et al., 2006)

demonstrated that immunogens other than alpha-toxins are important in protective

immunity against C. perfringens infection in broiler chickens.

17

CHAPTER III

CHARACTERIZATION OF CLOSTRIDIUM PERFRINGENS AND

CLOSTRIDIUM SEPTICUM ISOLATES FROM CELLULITIS CASES

18

Summary

Twelve isolates each of C. perfringens and C. septicum were isolated from

cellulitis lesions of turkeys from Minnesota. Characterization of these isolates were done

through in vitro laboratory assays including heat resistant spore count, hemolysis assay

and mice lethal studies as well as by genomic and proteomic studies. We observed

variations in the heat resistant spore counts and hemolytic titers between the isolates

examined. We also noticed differences in the MLD50 of C. perfringens and C. septicum

isolates. Clostridium perfringens strains isolated from cellulitis cases were of toxino-type

A by PCR. The phylogenetic tree constructed from C. perfringens isolates using MLST

indicated clustering among C. perfringens from cellulitis cases. However, two major

clusters were found among the C. perfringens isolates isolated from cellulitis cases. The

phylogenetic tree constructed from C. septicum isolates indicated a high level of

conservation present within the housekeeping gene fragments of C. septicum. The major

secretory toxins we identified in C. perfringens isolates from cellulitis cases by MALDI-

TOF mass spectrometry were phospholipase, collagenase, hyaluronidase, dnase, enolase,

muramidase, pyruvate kinase and hypothetical proteins. We observed two distinct

proteomic profiles for C. perfringens isolates. However, we observed only one type of

proteomic profile for C. septicum isolates. The major secretory toxins we identified in C.

septicum were alpha toxin, septicolysin, sialidase, Dnase, flagellin and hypothetical

proteins. More sequence types were observed among C. perfringens than among C.

septicum isolates. Our results indicate that C. perfringens isolates vary much in their

toxin expression and appeared more diverse. C septicum isolates were found to be less

diverse but more lethal than C. perfringens isolates. These data indicate that C. septicum

19

isolates are more likely to cause cellulitis mortality in turkeys than C. perfringens

however more studies are needed to substantiate this assertion.

20

INTRODUCTION

Clostridium perfringens and Clostridium septicum have been suspected in

playing a role in causing cellulitis and mortality in turkey population. Their role has been

suggested as early as 1939 (Fenstermacher et al., 1939) but conclusive studies were not

undertaken. There are many reports about isolation of C. perfringens from cellulitis or

gangrenous dermatitis cases in chickens (Helfer et al., 1969; Willooghby et al., 1996).

Organisms isolated from cellulitis lesions of turkeys in Minnesota during the years

2005- 2007 primarily include C. perfringens and C. septicum and either alone or in

combination. Clostridium perfringens alone was reported to cause cellulitis in turkeys and

experimental reproduction of the disease has been successful in adult turkeys (Carr et al.,

1996). But experimental reproduction of cellulitis with C. septicum is not yet reported in

chickens or turkeys. Though Clostridium perfringens and C. septicum were attributed as

the primary agents causing cellulitis in turkeys based on the isolation results, the role of

C. perfringens and C. septicum isolates in the development of cellulitis lesions was not

studied in detail.

Clostridium is one of the most widespread of all pathogenic bacteria affecting

humans and animals. Its main habitats are the soil and the intestinal tracts of animals and

man. Clostridium perfringens is found in the intestinal tract of healthy poultry as a normal

inhabitant, usually in low numbers (Songer, 1996). It has been isolated from processed

carcasses as well as from the processing plants (Craven et al., 2001b; Martel et al., 2004).

Hatcheries were identified as a potential source and reservoir for C. perfringens in

integrated poultry operations (Craven et al., 2001a).

21

Clostridium septicum is an organism of considerable medical and veterinary

importance. But little is known about the distribution and sources of C. septicum in poultry

production facilities.

The objective of this study was to characterize Clostridium perfringens and

Clostridium septicum isolates from cellulitis cases in turkeys. Characterization of the

isolates were done through in vitro laboratory studies like hemolysis assay and mice

lethal studies as well as by genomic and proteomic studies. This enabled us to select the

isolates for developing a disease model as well as for vaccine production.

MATERIALS AND METHODS

1. Source of Bacteria

Clostridial isolates included in this study originated from field cases of cellulitis

in turkeys either submitted to our laboratory from different turkey farms of a turkey

company in Minnesota or were collected by us during farm visits from 2007-2008.

2. Bacterial Isolation and Identification

Samples were collected in 0.1% Bacto peptone (Becton, Dickinson and Company,

Sparks, MD) using sterile swabs from cellulitis lesions during necropsy of fresh birds.

Isolation of Clostridium perfringens was done in cooked meat medium (Jong et al., 2002)

at 37 C for 24 h under anaerobic conditions in Mitsubhisi® anaerobic boxes (Mitsubhisi

gas chemical co, NY, USA). Oxygen absorbing gaspaks (Anaeropack-Anaero, Mitsubhisi

gas chemical co, NY, USA) were used in anaerobic boxes to maintain strict anaerobic

conditions. Subsequent subculturing was done on anaerobic sheep blood agar plates

22

(Oxoid, NY, USA) or tryptose sulphite cycloserine (TSC) agar with egg yolk (Merck,

Darmstadt, Germany) agar plates (Oxoid, NY, USA). They were examined for their

morphology, staining, and cultural characteristics. Specicity of C. perfringens colonies

was ascertained by their growth on tryptose sulphite cycloserine (TSC) agar with egg

yolk (Merck, Darmstadt, Germany) and reverse CAMP test.

Isolation of C. septicum was made in cooked meat medium and brain heart

infusion (BHI) broth (Oxoid, Ogdensburg, NY, USA) under anaerobic conditions.

Subsequent subculturing was done on anaerobic sheep blood agar plates (Oxoid, NY,

USA) or Phenyl ether alcohol (PEA) agar plates (Oxoid, NY, USA). They were examined

for their morphology, staining, and cultural characteristics.

Suspected Clostridium perfringens and Clostridium septicum isolates was further

identified by biochemical tests using API 20A, (Gubash, 1980; Buchanan, 1982) and

PCR. Clostridium perfringens was confirmed by 16S rDNA based PCR (Wang et al.,

1994) and was genotyped using multiplex PCR (Meer et al., 1997) and alpha-toxin PCR

(Kalender et al., 2005). Clostridium septicum was confirmed by alpha-toxin PCR (Sasaki

et al., 2001). Twelve isolates each of Clostridium perfringens type A (UMNCP 1 through

UMNCP 12) and Clostridiun septicum (UMNCS 101 through UMNCP 112) were

included in this study after confirming their identity.

For making the spore culture, Clostridium perfringens inoculated blood agar

plates was incubated at 37 C for 24 h under anaerobic conditions in Mitsubhisi®

anaerobic boxes (Mitsubhisi gas chemical co, NY, USA) to make the preculture stock.

One individual colony was picked and transferred to 10ml of pre-reduced Fluid

23

thioglycolate (FTG) medium (Difco) and incubated at 37 C for 24 h under anaerobic

conditions to make the preculture stock.

Sporulation properties of C. perfringens isolates were examined as previously

described (de-Jong et al 2002; Shih et al., 1996). Briefly, the isolates were cultured in 10

ml pre-reduced FTG tubes for 24 h at 37 C and subsequently they were subcultured in

tryptose sulphite cycloserine (TSC) agar with egg yolk (Merck, Darmstadt, Germany).

Inoculated plates were incubated anaerobically for 24 h. One individual colony was

inoculated into a preculture in FTG, which was incubated for 24 h and subsequently

subcultured into Duncan and Strong (DS) sporulation media (Duncan and Strong, 1968).

A 5 ml of 24h growth in FTG media was transferred to 30 ml DS media and incubated

under anaerobic condition for 24 h at 37C for maximum spore yield. Spore cultures of

Clostridium septicum were prepared by inoculating into fresh BHI and incubated under

anaerobic condition at 37C for 18 h and 25C for 6 h (Cortinas et al., 1997; McCourt et al.,

2005).

3. Estimation of heat resistant spore counts

Heat resistant spores from isolates of Clostridium perfringens and C. septicum

were estimated as described before (de Jong et al., 2002). Clostridium perfringens and

Clostridium septicum spore cultures were heated for 20 min at 70C in a water bath and

immediately cooled in ice water. Ten-fold serial dilutions were made from these spore

cultures in normal saline solution and pour plated into TSC agar plate for C. perfringens

and Phenylethyl alcohol blood agar (PEA) plates for C. septicum. The number of discrete

24

colonies was then counted after 12 h of anaerobic incubation at 37C to assess the titer of

viable spores present. The experiment was repeated four times.

4. Estimation of Toxin production in vitro by Hemolytic assay

Hemolytic activity for C. perfringens and C. septicum cultures was determined by

a microtiter assay as previously described (Ballard et al., 1992; Titball et al., 1989).

Briefly, 3 ml of sheep blood was centrifuged and the sedimented RBCs were washed 3

times with normal saline and a 1% RBC suspension in PBS without Calcium and

Magnesium was made. The culture supernatants were diluted two-fold serially across the

microtiter plate in 100 µl of PBS without Calcium and Magnesium. One-hundred µl of

1% sheep RBC suspension was added into all the wells and the microtiter plate was

incubated at 37C for 1h. The highest dilution of the C. perfringens or C. septicum culture

supernatant producing visible hemolysis was considered as the toxin titer. The

experiment was repeated four times.

5. In vivo studies on the biological effect of Clostridia and their toxins in mice

Twelve isolates each of C. perfringens and C. septicum were subjected to the

mice assay. All in vivo experiments involving Clostridia were conducted with prior

approved protocols from Institutional Animal Care and Usage Committee (IACUC) and

Institutional Biosafety Committee (IBC) of University of Minnesota and the procedures

were performed accordingly.

The 24 h spore cultures of both Clostridium perfringens and Clostridium septicum

were tested in mice for the evaluation of the potency of Clostridial toxins present in the

25

spore culture. The biological activity of both C. perfringens and C. septicum toxins

present in the culture were assessed by determining the 50% minimum lethal dose

(MLD50) in mice. MLD50 is described as the minimum amount of toxin required to cause

mortality in half of the mice population and is expressed in milligrams. The protein

content in the culture supernatants was quantitated using a Pierce BCA protein Assay kit

(Thermo scientific, IL, USA).

The MLD50 was determined as follows. Briefly, six serial dilutions from 100 ul of

each spore culture was made in sterile saline. Six mice (Swiss Webster outbred females

18 to 24 g) per group were injected intraperitoneally with 200ul of serially diluted C.

perfringens or C. septicum culture supernatants. Six mice served as sham-inoculated

controls and received an equal volume of sterile broth. The effect of Clostridial toxins in

mice includes marked respiratory distress, severe convulsions and death. Time to death

was monitored for each mouse to obtain an estimate of lethal fractions. The number of

survivors was recorded for each group after 48 h. The MLD50 was calculated by the

method of Reed and Muench (Reed et al., 1938). The experiment was repeated only once.

6. Multilocus sequence typing (MLST)

All twelve isolates each of C. perfringens and C. septicum isolated from clinical

cases of cellulitis were subjected to MLST. Clostridium perfringens and C. septicum

strains were grown on anaerobic sheep blood agar plates (Oxoid, Ltd). The plates were

incubated anaerobically at 37°C for 24 h using the AnaeroPack SystemTM (Mitsubishi

Gas Chemical America, Inc., New York, NY). Clostridium perfringens and C. septicum

DNA for PCR was prepared by boiling. Briefly, bacterial cells were scraped from plates,

26

resuspended in water and subjected to boiling for 20 min. Cell debris was removed by

centrifugation at 13,100g for 5 min and the supernatant was used as template DNA for

PCR.

Seven housekeeping loci and the alpha toxin gene were selected for the

characterization of C. perfringens isolates by MLST as described before (Jost et al.,

2006). They were ddlA (D-alanine-D-alanine ligase), dut (deoxyuridine-triphosphatase),

glpK (glycerol kinase), gmk (deoxyguanylate kinase), recA (recombinase), sod

(superoxide dismutase), tpi (triose phosphate isomerase) and plc (alpha toxin). The

choice of these genes was based on their use in MLST schemes before (Jost et al., 2006).

The C. perfringens primers used in this study are shown in Table 1.

Similarly, isolates of C. septicum were also subjected to PCR amplifications with

Platinum Taq DNA Polymerase (Invitrogen Life Technologies) using oligonucleotide

primer sequences for the house-keeping genes csa (alpha toxin), col A (collagenase),

recA (recombinase), ddl (D-alanine-D-alanine ligase), gyr A (DNA gyrase subunit A),

dnaK (dna kinase), groEL (GroEL protein) and glpK (glycerol kinase) as previously

described (Neumann et al., 2009). The C. septicum primers used in this study are showed

in Table 2.

PCR amplification of DNA was performed using Taq DNA Polymerase (Platinum

PCR Supermix, Invitrogen) in a reaction buffer containing 1.65mM MgCl2, 0.22mM

dNTPs and 1.5µM of each oligonucleotide primer. PCR was performed using 35 cycles,

with one cycle consisting of 1 min at 94C, 1 min at 50C and 1 min at 72C. Final

extension step was at 72C for 5 min. The PCR products were then purified using

Qiaquick PCR purification kit (Qiagen). The purified PCR products in duplicate were

27

submitted to Biomedical Genomic Center, University of Minnesota for Sanger

sequencing. The DNA sequences were edited using Sequencer version 4.1.2 (Gene

Codes, Ann Arbor, MI) and added to an alignment containing the genes extracted from

published genome sequences of other C. perfringens isolates.

A phylogenetic analysis was conducted using software MEGA 4.1

http://www.megasoftware.net/ mega41.html). The sequence quality of all the Trace files

in a chromatogram was read and poor quality sequences were trimmed. The sequence

files were renamed according to the genes and isolate names. The sequences were then

aligned using Mega 4.1 software and Polymorphisms were recorded. We included

sequence types from reference C. perfringens strains SM101, ATCC13124, Strain13

which originated from food poisoning, human gas gangrene and soil respectively and C.

septicum ATCC12464 in the dendrogram to determine the phylogenetic relationship

between them. No information is available regarding the origin of C. septicum

ATCC12464 strain.

Two-parameter distances were computed and used to generate phylogenetic trees

using neighbour-joining method. The statistical reliability of internal branches was

assessed from 500 bootstrap pseudoreplicates. Sequence types (STs) were assigned on the

basis of unique allelic profiles.

7. Two-dimensional gel electrophoresis (2-DiGE) analysis and MALDI-TOF mass

spectrometry.

All twelve isolates of C. perfringens and C. septicum mentioned in the previous

chapter were used in this study. The Clostridium perfringens culture from fluid

28

http://www.megasoftware.net/%20mega41.html

thioglycolate Medium (DifcoTM) after anaerobic incubation for 24 h at 37°C were

transferred to DS medium to enhance toxin production. The Clostridium septicum were

cultured in BHI medium (DifcoTM) after anaerobic incubation for 24 h at 37°C. The

culture supernatants were treated with protease inhibitors and dialyzed by use of 10-kDa

cutoff zeba filter columns (Millipore Inc., Billerica, MA) to obtain secreted proteins. The

protein content in the culture supernatants was quantitated using a Pierce BCA protein

Assay kit (Thermo scientific, IL, USA)

One hundred micrograms of protein sample was subjected to Immobilized pH

gradient (IPG) Isoelectric focusing (IEF) in the 1st Dimension. Readystrip IPG strips

(Bio-rad laboratories, Hercules, CA) 11 cm of PH range 3-10 were used in this study.

This separated proteins by their charge (pI). For SDS second dimension 8-16% Criterion

precast gels (Bio-rad laboratories, CA) were used. The resulting gel was then Deep

Purple dye (Thermo scientific, Rockford, IL) stained for visualizing the protein spots.

The gels were then scanned using an Eagle eye spectrophotometer to obtain image files.

The protein spots were identified, spot picked and trypsin digested before subjecting to

MALDI-TOF mass spectrometry analysis. Two sets of spots were picked for each isolate

and identified by MALDI-TOF mass spectrometry analysis.

Statistical analysis: Statistical analysis was performed using SAS software (SAS

Language Version 9.2, SAS Institute Inc, Cary, NC). Pearson correlation coefficient was

used to determine whether there is any correlation between spore count, hemolytic titers

and MLD50. A P value of less than 0.05 was considered significant.

29

RESULTS

1. Bacterial Isolation and Identification

Clostridium perfringens appeared as large smooth convex colonies having a

double-zone of β-hemolysis on anaerobic sheep blood agar. They were positive for

reverse CAMP test and were identified as Clostridium perfringens based on API-20A

testing. All the twelve Clostridium perfringens isolates were identified as Clostridium

perfringens type A based on multiplex PCR and alpha-toxin PCR.

Clostridium septicum produced a thick swarming growth with a narrow zone of β-

hemolysis on anaerobic sheep blood agar. All the twelve Clostridium septicum isolates

were confirmed using API 20A test and alpha-toxin PCR for C. septicum.

2. Estimation of heat resistant spore counts:

We observed a variation in the heat resistant spore counts between the isolates

examined (Tables 3 and 4). Among twelve isolates of Clostridium perfringens, the

average spore count ranged from 1.4 x 108 to 5.2 x 106 spores/ml. Two isolates UMNCP

01 and UMNCP 04, recorded a maximum spore count of 1.4 x 108 and 6.4 x 107

spores/ml respectively. Among the C. septicum isolates examined, the spore count ranged

from 4.8 x 107 to 2.6 x 105 spores/ml. The isolates UMNCS 106 and UMNCS 107 had

the highest spore count of 4.8 x 107 spores/ml and 8.2 x 106 spores/ml respectively.

3. Estimation of Toxin production in vitro by Hemolytic assay

We found differences in the hemolytic titers for C. perfringens spore cultures.

They ranged from 512 to 4 (Tables 3 and 4). The hemolytic titer for C. septicum spore

30

cultures ranged from 256 to 64. Clostridium perfringens isolate UMNCP 01 and UMNCP

04 had the highest titer of 512. The hemolytic titers for C. septicum spore cultures were

256 and 128 for isolates UMNCS 106 and UMNCS 107, respectively. Clostridium

septicum isolate UMNCS 106 had the highest titer of 256.

4. In vivo studies on the biological effect of Clostridia and their toxins in mice

Mice assay studies revealed a difference between C. perfringens and C. septicum

isolates in their MLD50 values (Tables 3 and 4). The MLD50 of most potent C.

perfringens spore cultures were found to be 2.12 mg for UMNCP 01 and 2.75 mg for

UMNCP 04. For most C. septicum isolates (UMNCS 106 and UMNCS 107) the MLD50

was 0.024 mg and 0.031 mg, respectively.

Among C. perfringens strains examined, correlation was found between heat

resistant spore count and hemolytic titers (P = 0.0004) but the correlation was less

between hemolytic titers and MLD50 values (P = 0.04). There was no correlation

observed between spore count and MLD50 values (P = 0.11).

Among C. septicum strains examined, correlation was found between heat

resistant spore count and hemolytic titers (P = 0.003) but the correlation was less between

hemolytic titers and MLD50 values (P = 0.02). There was no correlation observed

between spore count and MLD50 values (P = 0.06).

31

5. To determine the genetic relationship among C. perfringens and Clostridium

septicum isolates from cases of cellulitis in turkeys using Multilocus sequence typing

(MLST)

Clostridium perfringens strains isolated from cellulitis cases were of toxino-type

A by PCR. Seven housekeeping genes and one toxin gene were analyzed for

characterization of both C. perfringens and C. septicum isolates by MLST. MLST

analysis of 12 C. perfringens isolates identified 11 sequence types. Two major clusters

were found among the C. perfringens isolates examined (Figure 1). All the C. perfringens

isolates examined fell into a single cluster except UMNCP 8 and UMNCP 9 isolates

which clustered with C. perfringens strain 13.

Some of the C. perfringens isolates like UMN CP 05 and UMNCP06 with less

hemolytic titers fell into a cluster by itself. Besides that no such correlation between toxin

production and genetic relatedness was observed. Members in both clusters varied

considerably in their hemolytic toxin titer representing no direct connection between the

expression of hemolytic toxin and phylogenetic relationship. Among the eight loci

examined plc locus had the maximum number of alleles (8).

MLST analysis of 12 C. septicum isolates yielded only two sequence types (ST).

All the C. septicum isolates examined fell into a single cluster along with ATCC strain

12464 except UMNCS9 isolate (Figure 2). Two polymorphic sites were identified in the

amplified fragment of gyrA.

32

6. To identify the secretory components of C. perfringens and Clostridium septicum

using 2-DiGE analysis and MALDI-TOF mass spectrometry.

The major secretory toxins we identified in C perfringens isolates by MALDI-

TOF mass spectrometry were phospholipase, collagenase, hyaluronidase, Dnase, enolase,

muramidase, pyruvate kinase and hypothetical proteins. We observed two distinct

proteomic profiles for C. perfringens isolates (Figure 3). We observed only one type of

proteomic profile for C. septicum isolates (Figure 4). The major secretory toxins we

identified in C. septicum isolates were alpha toxin, septicolysin, sialidase, Dnase,

Transferrin, flagellin and Gelsolin precursor (actin depolymerizing factor).

DISCUSSION

In chickens, a condition similar to cellulitis, often referred to as gangrenous

dermatitis, is found to be caused by a different set of bacterial pathogens like Escherichia

coli (Frazier, 1964; Kumor et al., 1998; McDevitt et al., 2006), Clostridium perfringens

and Clostridium novyi (Reed et al., 1938; Smith-Slatas et al., 2006), Clostridium septicum

(Gomis et al., 1997; Wilder et al; 2001) and Staphylococcus aureus (Wang et al., 1994).

Escherichia coli, Streptococcus fecalis, Proteus sp., Bacillus sp., Staphylococcus aureus

and Clostridium septicum were isolated from gangrenous dermatitis cases in broiler

chickens (Fenstermacher et al., 1939). However organisms isolated from the cellulitis

lesions of turkeys in Minnesota in the recent years primarily include C. septicum and C.

perfringens either alone or in combination.

Most of the isolates of C. perfringens and C.septicum that produced higher spore

counts also showed higher hemolytic titers. This is in agreement with the earlier findings

33

that alpha-toxin production is associated with sporulation in Clostridium perfringens

(Varga et al., 2004). However, Cortinas et al., 1997, reported an absence of correlation

between hemolytic titers and cell growth with few C. septicum isolates tested. We

observed no correlation between spore counts and MLD50 values either in C. perfringens

or C.septicum isolates. This could be due to the differences in the production of non-

hemolytic toxins. The low hemolytic activity could be due to the action of endogenous

proteases on the hemolytic toxins produced by C. septicum (Cortinas et al., 1997).

Though the spore count and hemolytic effects of C. perfringens were found to be

higher than C. septicum in vitro, the lethal effects in mice as well as in turkeys were not

comparable. The marked hemolytic activity of C. perfringens isolate may be due to the

secretion of alpha-toxin (phospholipase C) which is markedly hemolytic (Timoney et al.,

1998). Hemolytic activities have also been described for the alpha-toxin produced by C.

septicum. However, the alpha-toxin of C. septicum is not related to the alpha-toxin of C.

perfringens since it does not exhibit phospholipase C activity and does not cross-react

immunologically with C. perfringens alpha-toxin (Amimoto et al., 2002). The mortality

studies in mice showed that C. septicum was more toxic than C. perfringens isolates.

We observed differences in the toxicity and biological effects of different strains of

C. perfringens and C. septicum either in vitro or in vivo in mice. Differences in the toxin

production ability as well as disease causing potential between Clostridial isolates have

been reported before. In an experimental infection with Clostridia, all the five isolates of

C. perfringens and three of four of C. septicum isolates did not produce any cellulitis

lesions in 10-week-old turkey breeder hens when given intravenously (Tellez et al., 2009)

whereas only one isolate of C. septicum caused disease.

34

Most of the C. perfringens isolates examined fell in a single cluster indicating a

clonal relationship. Among the eight loci examined plc locus had the maximum number

of alleles (8) corroborating the fact that plc is highly diverse in nature genetically

(Chalmers et al., 2008). The phylogenetic tree constructed from C. perfringens isolates

indicated a high level of diversity within the housekeeping gene fragments of C.

perfringens. Our findings are in agreement with previous reports (Jost et al., 2006).

However, the dN/dS ratio indicated that most of these polymorphisms found in the eight

genes examined resulted from synonymous substitutions. This indicates that no

functional change occurred in any of the final protein products.

The data generated by MLST indicate that there is considerable genetic diversity

among C. perfringens isolates. Some of the C. perfringens isolates like UMNCP05 and

UMNCP06 with less hemolytic titers fell into a cluster by itself. Clostridium perfringens

isolates UMN CP05 and UMNCP06 were found to have the least hemolytic titers.

Besides that no such correlation between toxin production and genetic relatedness was

observed. Members in both clusters varied considerably in their hemolytic toxin titer

representing no direct connection between the expression of hemolytic toxin and

phylogenetic relationship.

UMNCP 8 and UMNCP 9 isolates were found to be clustered with reference

isolate C. perfringens strain 13. The latter is reported to be a natural isolate from the soil

and pathological studies have shown that this strain can establish experimental gas

gangrene (myonecrosis) in a murine model (Shimizu et al., 2001; Rood, 1998).

Clostridium perfringens strain ATCC13124 which is a human gas gangrene isolate and

SM101, a human food poisoning isolate fell apart in a distant cluster indicating no

35

phylogenetic relationship with any of the turkey isolates examined. There are studies

which reveal that C. perfringens from porcine origin and NE cases were clonal (Chalmers

et al., 2008). However, there were no reports regarding C. perfringens isolates from

cellulitis cases in turkeys.

The present study indicates that C. perfringens isolates from cellulitis cases

showed substantial clonality. All the reference strains fall in a subset of its own. This

indicates that C. perfringens from cellulitis cases showed no ancesteral relationship with

reference strains like ATCC 13124 and SM 101.

In summary, MLST results indicate that there is considerable genetic diversity

among C. perfringens isolates from cellulitis cases and that the MLST scheme can be

applied for typing of C. perfringens isolates causing cellulitis. MLST analysis of C.

septicum isolates from cellulitis cases appeared highly clonal. The diversity observed

among isolates of C. septicum in this study is considerably less than has been reported for

similar analyses performed on C. perfringens (Jost et al., 2006) and C. septicum

(Neumann et al., 2010). One possible explanation for this is that our analysis included

primarily isolates from cellulitis cases in turkeys whereas these other two studies

examined isolates from many different animal hosts and disease presentations.

The major secretory toxins in C. perfringens isolates were phospholipase,

collagenase, hyaluronidase, dnase, muramidase and hypothetical proteins. Hypothetical

proteins were observed mostly in profiles of C. perfringens isolates having highest

hemolytic titers. The use of it as a biomarker for cellulitis causing C. perfringens isolates

needs to be further investigated. Clostridium perfringens is reported to produce different

types of toxins and the toxin profile varies with the strain of bacteria (Sawires et al. 2006;

36

Smedley et al., 2004). C. perfringens produces extracellular toxins including beta2 toxin,

enterotoxin, perfringolysin, collagenase, lambda toxin, hyaluronidase, dnase,

neuraminidase and urease (Sawires et al. 2006). Our results suggest involvement of

different toxins of C. perfringens in pathogenesis of cellulitis in turkeys.

Clostridium perfringens isolate UMNCP01 that appeared most potent differed in

their secretory protein profile from less potent isolate like UMNCP06. The role of

hypothetical protein 1232 needs to be further investigated. The functional annotation of

this protein is suggestive of a protease. The proteomic profiles of all C. septicum isolates

appeared identical.

The major secretory toxins in C. septicum isolates were alpha toxin, septicolysin,

sialidase, Dnase, flagellin and Gelson precursor (actin depolymerizing factor). The

Gelson precursor (actin depolymerizing factor) might have some role in the pathogenesis

of cellulitis, but more studies are warranted to confirm this assertion. Since C. septicum

genome is not completely sequenced, not much information is available for the

development of a recombinant vaccine at this point. Unlike C. perfringens we observed

only one type of proteomic profile for C. septicum isolates. These results support our

findings in the MLST analysis as well as previous reports (Neumann et al., 2010) that

genomes of C. septicum poultry isolates are highly conserved.

37

TABLES

Table 1. Oligonucleotide primers used for C. perfringens MLST (Jost et al., 2006)

ene Primer Sequence (51–31) Amplicon size

(bp)

plc Forward ATATGAATGGCAAAGAGGAAAC 544

Reverse AGTTTTTCCATCCTTTGTTTTG

ddlA Forward ATAATGGGGGATCATCAGTTGC 429

Reverse TTATTCCTGCTGCACTTTTAGG

dut Forward TTAAGTATTTTGATAACGCAAC 441

Reverse CTGTAGTACCAAATCCACCACG

glpK Forward TGGGTTGAGCATGATCCAATGG 547

Reverse CACCTTTTGCTCCAAGGTTTGC

gmk Forward TAAGGGAACTATTTGTAAAGCC 475

Reverse TACTGCATCTTCTACATTATCG

recA Forward GCTATAGATGTTTTAGTTGTGG 475

Reverse CTCCATATGAGAACCAAGCTCC

sod Forward GATGCTTTAGAGCCATCAATAG 475

Reverse AATAATAAGCATGTTCCCAAAC

tpi Forward AAATGTGAAGTTGTTGTTTGCC 451

Reverse CATTAGCTTGGTCTGAAGTAGC

38

Table 2. Oligonucleotide primers used for C. septicum MLST (Neumann et al., 2009)

Gene Primer Sequence (51–31) Amplicon size

(bp)

colA Forward GCTTCAGCATATGAACTAGTTAAGGG 562

Reverse CTGGATGTCCTTCGTCTAAAGGCT

csa Forward TGATTGCAATTCAGTGTGCGGCAG 632

Reverse GCTTGTCTCAACTTTACTGTCAGC

ddl Forward GCTTTGCATGGTAAATTTGGTGAAGATGG 597

Reverse AGGAGTCCTATCCTCTGTAACTATCA

dnaK Forward GCAGTTATGGAAGGTGGAGAACCA 744

Reverse ACCAGTTGCATCTGCAGTGATGAG

glpK Forward CCCAATGGAAATTTGGGCAAGCCA 606

Reverse GCTGCTTGTTGGTCTCCAGCAATA

groEL Forward GGGAGCACAGTTAGTGAAGGAAGT 630

Reverse CCTGGTGCCTTTACTGCAACACAT

gyrA Forward GGTGATAATGCAGCGGCAATGAGA 792

Reverse GATATGAGCTCTTGCTTCAGCC

recA Forward TGCAGAGCATGCCTTAGATCC 600

Reverse CCATATGAGAACCAAGCTCCAC

39

Table 3. Heat resistant spore count, hemolytic titers and MLD50 of various isolates of C.

perfringens

Isolate ID# source HR-spore count Hemolytic titer MLD50

(per ml) (mg)

1 UMNCP 01 skin lesion 14 x 107 512 2.12

2 UMNCP 02 skin lesion 0.84 x 107 224 4.86





7 UMNCP 07 liver 5.8 x 107 144 5.64


9 UMNCP 09 liver 4.6 x 107 96 5.68




40

Table 4. Heat resistant spore count, hemolytic titers and MLD50 of various isolates of C.

septicum

Isolate ID# source HR-spore count hemolytic titer

MLD50

(per ml) (mg)

es n

12 UMNCS 112 skin lesion 6.4 x 106 112 0.168

1 UMNCS 101 skin lesion 3 .9 x 106 96 0.12


3 UMNCS 103 skin lesion 4 x 106 112 0.14






9 UMNCS 109 skin l io 0.26 x 106 96 0. 236

10 UMNCS 110 liver 5.1 x 106 80 0.24


41

d

olates from cellulitis cases and reference isolate

Strain 13, ATCC13124 and SM101.

FIGURES

Figure 1. Phylogenetic tree constructed by Neighbor-Joining method from concatenate

sequences of various C. perfringens is

42

Figure 2. Phylogenetic tree constructed by Neighbor-Joining method from concatenated

sequences of various C.septicum isolates from cellulitis cases and reference isolate

ATCC12434.

43

Figure 3. Two distinct proteomic profiles of C. perfringens UMNCP01 and UMNCP 06

isolates after 2-DiGE.

44

Figure 4. Proteomic profile of of C. septicum after 2-DiGE

45

CHAPTER IV

ROLE OF CLOSTRIDIUM PERFRINGENS AND CLOSTRIDIUM SEPTICUM IN

CELLULITIS IN TURKEYS

46

Summary

Clostridium perfringens and Clostridium septicum are widespread in the poultry

environment and those happened to be the primary agents we isolated from cellulitis

cases in turkeys. The objective of this study was to examine the role of Clostridium

perfringens and Clostridium septicum in the experimental development of cellulitis

lesions in turkeys. The study was conducted in turkeys of 3-weeks of age and 7-weeks of

age. A higher mortality was observed in turkey poults given C. septicum culture as

compared to birds given C. perfringens culture in both age groups of birds. These

observations were comparable to our mice assay results where C. septicum culture was

found to be more lethal to mice even at lower doses when compared with the C.

perfringens culture. However, mortality and cellulitis lesions were more pronounced in

birds of 7-weeks of age than in birds of 3-weeks of age. Surprisingly, in 7-week-old

birds, the development of fulminant cellulitis lesions and mortality was markedly higher

than in 3-week old birds. Administration of either 1 ml of C. septicum or 3 ml of C.

perfringens produced a higher percentage of cellulitis mortality in 7-week-old birds. The

cellulitis lesion development was also more prominent in 7-week old birds than in 3-

week old birds. More over, there were no C. perfringens or C. septicum noticed

microscopically in the muscle tissues examined from lesions in 3-week old birds. In 7-

week old birds died of cellulitis, there were numerous multiple patches of muscles in

which the fibers were fragmented because of bacterial multiplication giving the areas a

“moth eaten” appearance. There were several areas in the subcutaneous tissues and

muscles that were infiltrated by large numbers of inflammatory cells. Multiplication of

either C. perfringens or C. septicum in the muscle tissues of cellulitis affected birds has not

47

been reported before. Numerous thick, rod-shaped bacteria were also present in the areas

of inflammation in the subcutaneous tissues and muscle fibers clearly indicating the

multiplication of bacteria within these regions and subsequent damage due to the

enormous amounts of toxins liberated. Thus, C. septicum appears to be more potent in

causing severe cellulitis lesions and higher mortality in turkeys although the role of C.

perfringens in causing fatal cellulitis cannot be ignored. Our results indicate that both

agents may contribute to the outcome of the disease. Our cellulitis disease model offer

promise to use it as a challenge model in the development of vaccines against cellulitis in

turkeys.

48

INTRODUCTION

Clostridium perfringens and Clostridium septicum have been suspected in playing

a role in causing cellulitis and mortality in turkeys. Organisms isolated from the cellulitis

lesions of turkeys in Minnesota primarily include C. perfringens and C. septicum either

alone or in combination. The isolation of C. perfringens from cellulitis cases of chickens

and turkeys has been emphasized in many reports (Carr, 1996; Hofacre et al., 1986). Little

is known about the pathogenic potential of C. perfringens and C. septicum as disease

reproduction is often difficult. Clostridium perfringens was reported to cause cellulitis in

turkeys and experimental reproduction of the disease has been successful in adult turkeys

(Carr et al., 1996). Detailed studies had not been under taken with C. septicum so far.

Recent reports claim C. septicum as the primary agent causing cellulitis in turkeys based

on the isolation results (Tellez et al., 2009), however the role of C. septicum isolates in

the development of cellulitis lesions was not studied in detail.

Clostridium perfringens and Clostridium septicum are widespread in the poultry

environment. Clostridium perfringens is even found in the intestinal tract of healthy

poultry as a normal inhabitant, usually in low numbers (Songer, 1996). However, little is

known about the distribution and sources of C. septicum in poultry production facilities.

The age range at which cellulitis is seen in turkeys appears to be varied. In the

field, one can observe cellulitis cases even in birds as young as 7-weeks of age. Since

there is no consistency, we planned on examining birds of two different age groups: at 3-

weeks age and at 7-weeks of age. The younger age was first examined to see whether we

could reproduce the disease in this age group assuming that if this can be done it would

reduce housing, ease of handling considerations. The objective of this study was to

49

examine the roles of Clostridium perfringens and Clostridium septicum in the

experimental development of cellulitis lesions in turkeys.


Bacteria: The spore culture preparations of C. perfringens (UMNCP 01) and C. septicum

(UMNCS 106) which had the highest spore count, hemolytic activity and MLD50 were

used to study their effects in turkey poults. Clostridium perfringens culture containing 6.4

x 107 heat resistant spores/ml and C. septicum (UMNCS 106) culture containing 8.2 x 106

heat resistant spores/ml were used in this study. The MLD50 /ml of C. perfringens

(UMNCP 01) spore culture was found to be 2.12 mg and C. septicum (UMNCS 106)

0.024 mg.

Effect of C. perfringens and C. septicum in three-week-old turkeys

A dose-response study was conducted to examine the lethal effects of C.

perfringens and C. septicum spore cultures following subcutaneous inoculation and to

examine the development of any cellulitis in 3-week-old turkey poults. All experimental

protocols were approved by the Institutional Animal Care and Use Committee (IACUC)

and the procedures were performed in accordance with the requirements. The

experimental birds were kept and reared in Research Animal Resources (RAR) isolation

facilities of University of Minnesota.

Experimental design

The studies were conducted with seventy-two, 3-week-old commercial Nicholas

White turkey poults obtained from farms with no history of cellulitis. Two treatments

50

were done by subcutaneous injection with either C. perfringens or C. septicum spore

culture to assess the biological effect.

Briefly, the birds were randomly divided into two groups, I and II, of thirty six

birds each. Each group was further divided into six subgroups of six birds each. Birds in

each subgroup in group I were given different doses (0.5 ml, 0.75 ml, 1 ml, 2 ml and 3

ml) of C. perfringens culture alone subcutaneously on the breast region. Birds from group

II, were given different doses (0.5 ml, 0.75 ml, 1 ml, 2 ml and 3 ml) of C. septicum

culture alone subcutaneously in the breast region. One subgroup consisting of six birds

each in groups I and II served as sham-inoculated controls where 3ml of sterile anaerobic

culture media was injected subcutaneously on the breast region. The dead birds at 24 h

and 48 h were necropsied and tissue samples from cellulitis lesions were examined for

bacterial isolation and histopathology.

Effect of C. perfringens and C. septicum in seven-week-old turkeys

A dose-response study was conducted to examine the development of cellulitis in

7-week-old turkey poults. From most field observations, seven weeks of age appears to

be the earliest age at which turkeys were found to be developing cellulitis. In this study,

we used the same C. perfringens and C. septicum spore cultures that we used to test the

3-week-old poults. The inoculum contained either C. perfringens or C. septicum spore

cultures.

A study was conducted with 36, seven-week-old birds. Different doses of 0.5 ml,

0.75 ml, 1 ml, 2 ml and 3 ml each of a preparation containing either C. perfringens or C.

septicum spore culture were inoculated into three birds in each subgroup. The route of

51

inoculation in seven-week-old turkey poults was subcutaneously on the breast region.

The dead birds at 12 h, 24 h and 48 h were necropsied and tissue samples from cellulitis

lesions were examined for bacterial isolation and histopathology.

RESULTS

Effect of C. perfringens and C. septicum in three-week-old turkeys

The mortality observed in birds injected with spore cultures of C. perfringens or

C. septicum preparations is presented in Table 1. Higher mortality was observed with C.

septicum culture than with C. perfringens. Among birds inoculated with cultures from C.

septicum there was mortality even at lower doses (2 ml and 1 ml). In group of birds given

C. perfringens culture, mortality was seen only in those given a higher dose (3 ml). These

observations were comparable to our mice assay where the C. septicum culture was found

to be more lethal to mice even at lower doses when compared with the C. perfringens

culture.

A visual inspection revealed no blisters or vesicles in any part of the body

externally. The breast, thigh muscles and internal organs appeared hyperemic in birds.

There was little subcutaneous accumulation of inflammatory frothy fluid at the inoculated

area or anywhere else in the subcutaneous tissues.

Histopathology: Histopathology of the subcutaneous and muscle tissues from dead birds

revealed few microscopic lesions and inflammatory changes. These birds revealed few

localized areas of hemorrhage in the dermis and subcutis. The subcutis and deep dermis

were infiltrated by few heterophils and macrophages. A few thick rod-shaped bacteria

52

were visible in the inflamed subcutaneous area. The underlying muscle tissues did not

exhibit any inflammatory changes or presence of bacteria in any of the birds.

Effect of C. perfringens and C. septicum in seven-week-old turkeys

The mortality observed in birds injected with C. perfringens and C. septicum in

seven-week-old birds is shown in Table 2. We observed 100% mortality within 12 h

when either 2 ml or 3 ml of C. septicum spore culture was inoculated. The mortality was

66 % even in the one ml inoculated group. No mortality was observed when either 0.75

ml or 0.5 ml of the same inoculum was injected. However, in the C. perfringens

inoculated group, mortality was observed only with 3 ml of inoculation.

Cellulitis lesions developed mainly on the breast region and extended to the

abdomen and tail regions of the birds in both C. perfringens and C. septicum inoculated

groups (Figure 1A). Blisters and vesicles were also noticed on the breast region in both

groups (Figure 1B). In the C. perfringens treated group, 3 ml inoculation was required to

produce consistent lesions. There was subcutaneous accumulation of a frothy viscous

inflammatory fluid in cellulitis lesions in all the dead birds in both C. perfringens as well

as C. septicum inoculated birds (Figure 1C). However blisters and vesicles were more

prominent and extensive with birds given 2 ml or 3 ml of spore culture of C. septicum. In

birds given 2 ml of spore culture of C. septicum, the lesions were noticed on the breast

region as well as on the abdomen and tail regions. The presence of frothy sanguinous

subcutaneous exudate in the subcutaneous tissue was more apparent in C. septicum group

inoculated with 2 ml as compared to the 3 ml of C. septicum inoculated group or the 3 ml

of C. perfringens inoculated group. Palpitation of the affected areas often revealed

53

crepitation due to gas bubbles in the subcutis and musculature in all the dead birds. The

breast muscle and internal organs were hyperemic in birds given higher doses (1 ml, 2 ml

and 3 ml) of C. septicum.

Clostridium perfringens and C. septicum were reisolated from cellulitis lesions of

dead birds. All birds which were given 1 ml, 0.75 ml, and 0.5 ml of C. perfringens or

0.75 ml, and 0.5 ml of C. septicum survived without showing any signs of cellulitis. In

these birds, a visual inflammatory swelling was noticed at the site of inoculation in 24

and 48 h, which started subsiding by 96 h.

Histopathology: Histopathology of subcutaneous and muscle tissues from birds that

died of C. perfringens infection showed edema in the dermis. The dermal capillaries were

engorged with blood. There was marked fibrin exudation and acute hemorrhage in the

dermis and subcutis. The subcutis and deep dermis were infiltrated by large numbers of

heterophils and macrophages. Subcutaneous tissues contained large clear spaces

consistent with appearance of gas bubbles. Moderate numbers of thick rod-shaped

bacteria were visible in the inflamed area (Figure 1D). Many of the bacteria had terminal

spores. Some of the underlying muscle fibers had lost cellular detail (Figure 1F). There

were patches of muscle in which the fibers were fragmented giving the areas a “moth

eaten” appearance. There were a few areas in the muscle that were infiltrated by

inflammatory cells. Thick rod-shaped bacteria were present in the areas of inflammation

in the muscle fibers clearly indicating the multiplication of bacteria within the muscle

fibers and subsequent damage possibly due to the toxins liberated.

54

In the C. septicum inoculated group, subcutaneous and muscle tissue examination

revealed microscopic lesions and inflammatory changes in birds given higher doses (1ml,

2 ml and 3 ml) of C. septicum. These birds also showed edema in the dermis. The dermal

capillaries also were engorged with blood. There was marked fibrin exudation and acute

hemorrhage in the dermis and subcutis. In the 1 ml C. septicum inoculated group, the

subcutis and deep dermis were infiltrated by numerous heterophils and macrophages

where few to moderate bacteria were present (Figure 1E). However, in the 3 ml

inoculated group, the subcutaneous tissues contained large clear spaces consistent with

appearance of gas bubbles. Numerous rod-shaped bacteria were also present in the

inflamed area but the heterophils and macrophages were absent in areas where active

bacterial multiplication was going on.

In all the birds given higher doses (1ml, 2 ml and 3 ml) of C. septicum, the

underlying muscle fibers had lost considerable cellular detail (Figure 1G). There were

numerous multiple patches of muscle in which the fibers were fragmented giving the

areas a “moth eaten” appearance (Figure 1H). There were several areas in the muscle

that were infiltrated by large numbers of inflammatory cells. Numerous thick, rod-shaped

bacteria were also present in the areas of inflammation in the muscle fibers clearly

indicating the multiplication of bacteria within the muscle fibers and subsequent damage

possibly due to the toxins liberated.

DISCUSSION

We were able to successfully reproduce cellulitis lesions and mortality in turkeys.

Both Clostridium perfringens and Clostridium septicum were found to cause similar

55

cellulitis lesions in seven-week-old turkeys. In a recent report of an experimental

infection with Clostridia, all five isolates of C. perfringens and three isolates of C.

septicum out of four isolates tested did not produce mortality in 10-week-old turkey

breeder hens when given intravenously (Tellez et al., 2009). The reported mortality with

the C. septicum challenge was 78.5 % in a group of fourteen poults..

It is widely presumed that the natural outbreaks of cellulitis in turkeys are

associated with a proliferation of Clostridial pathogens in the poultry environment. One can

argue that a reasonable model of the disease should closely resemble the probable route of

natural pathogen exposure, i.e.: oral. Necrotic enteritis is yet another disease in chickens

caused by C. perfringens. Interestingly, efforts for consistent reproduction of NE by oral

inoculation of C. perfringens in different labs has resulted in extremely variable results

ranging from severe clinical signs, subclinical NE in exposed birds to no lesions at all

(Aiello et al., 2003, Carr et al., 1996; Olkowski et al., 1999 and Titball et al., 1999). Such a

significant lack of consistency in scientific data generated in different labs is puzzling in

view of a large number of claims identifying predisposing factors (Martel et al., 2004;

Ninomiya et al., 1994).

Attempts to reproduce cellulitis in turkeys by injecting purified alpha-toxin of C.

perfringens subcutaneously have been successful but injections with C. perfringens alone

have not been successful (Carr, 1996). Moreover, oral challenges have not been successful

with either C. perfringens or C. septicum in our repeated earlier attempts. This directed us

to investigate this disease in turkeys through subcutaneous inoculations of varying amounts

of spore cultures of C. perfringens and C. septicum.

56

The mortality studies in mice as well as in three-week old birds showed that C.

septicum spores cultures were more toxic than C. perfringens spore cultures. C.

perfringens spore cultures produced mortality only at higher doses of 3 ml whereas C.

septicum caused mortality even at the lower dose of 1 ml in turkeys.

Surprisingly, in seven-week-old birds, the development of fulminant cellulitis and

mortality was markedly higher than in three-week old birds. Administration of either 1 ml

of C. septicum or 3 ml of C. perfringens produced a higher percentage of mortality in

seven-week-old birds. The cellulitis lesion development was also more prominent in

seven-week old birds than in three-week old birds. More over, there were no C.

perfringens or C. septicum noticed in the muscle tissues examined from lesions in three-

week old birds. Considering the fact that a seven-week-old poult will weigh

approximately four times that of a three-week-old poult, it appears that the younger birds

are least susceptible to Clostridial cellulitis than the older birds. This finding also

correlates with the potential age of onset of cellulitis one may notice in turkeys in the

field.

The gross lesions produced by C. perfringens and C. septicum were almost

identical. However, the fulminant nature and the extent of lesions in the muscle tissue

were more pronounced in the C. septicum inoculated birds than the C. perfringens

inoculated birds. The gross and microscopical lesions obtained after administration of

spore cultures of either C. perfringens or C. septicum in seven-week-old birds were

comparable to the cellulitis lesions reported either with C. perfringens experimental

infection or field observations reported by Carr et al., 1996. However, multiplication of

57

either C. perfringens or C. septicum in the muscle tissues of cellulitis affected birds has not

been reported before.

In conclusion, C. septicum appears to be more potent in causing severe cellulitis

lesions and higher mortality in turkeys although the role of C. perfringens in causing fatal

cellulitis cannot be ignored. It could be possible that both agents may contribute to the

outcome of the disease. Our cellulitis disease model offer promise to use it as a challenge

model in the development of vaccines against cellulitis in turkeys.

58

TABLES

Table 1: Mortality in three-week old birds following subcutaneous injection with either

C. perfringens or C. septicum spore culture.

Group I A (C. perfringens) Group II A (C. septicum)

3ml 2ml 1ml 0.75ml 0.5ml 3ml 2ml 1ml 0.75ml 0.5ml

24h 1 0 0 0 0 3 2 1 0 0

48h 1 0 0 0 0 1 0 0 0 0

Total (%) 2(33) 0(0) 0(0) 0(0) 0(0) 4(66) 2(33) 1(16) 0(0) 0(0)

A Each dose was inoculated in 6 birds each (n=6)

59

Table 2: Mortality in seven-week old birds following subcutaneous injection with either

C. perfringens or C. septicum spore culture.

Group I A (C. perfringens) Group II A (C. septicum)

3ml 2ml 1ml 0.75ml 0.5ml 3ml 2ml 1ml 0.75ml 0.5ml

12h 1 0 0 0 0 3 0 0 0 0

24h 1 0 0 0 0 0 2 0 0 0

48h 0 0 0 0 0 0 1 1 0 0

Total (%) 2(66) 0(0) 0(0) 0(0) 0(0) 3(100) 3(100) 1(33) 0(0) 0(0)

A Each dose was inoculated in 3 birds each (n=3)

60

FIGURES

Figure 1. Gross lesions and histopathologic changes of subcutaneous and muscle tissue

in cellulitis in turkeys. Sections were stained with H& E. (A) Inflammation extending

from the breast region to abdomen and tail regions in a bird inoculated with 3 ml of C.

perfringens spore culture. Bar = 1cm. (B) Vesicles and blisters extending on the breast

region in a bird inoculated with 2 ml of C. septicum spore culture. Bar = 1 cm. (C) The

presence of froathy sanguinous exudate in the subcutaneous tissue in a bird inoculated

with 3 ml of C. perfringens spore culture. Bar = 1 cm. (D) Subcutaneous tissue showing

presence of moderate number of thick rod-shaped bacteria in a bird inoculated with 3 ml

of C. perfringens spore culture. Bar = 100 µm.(E) Subcutaneous tissue showing presence

of thick rod-shaped bacteria in a bird inoculated with 1 ml of C. septicum spore culture.

Bar = 100 µm. (F) Breast muscle studded with thick rod-shaped bacteria in a bird

inoculated with 3 ml of C. perfringens spore culture. Bar = 50 µm. (G) Breast muscle

studded with thick rod-shaped bacteria in a bird inoculated with 2 ml of C. septicum

spore culture. Bar = 50 µm. (H) Breast muscle with multiple patches of muscle tissue in

which the fibers were fragmented giving a “moth eaten” appearance in a bird inoculated

with 3 ml of C. septicum spore culture. Bar = 50 µm.

61

62

CHAPTER V

LABORATORY STUDIES ON A BIVALENT VACCINE CONTAINING

CLOSTRIDIUM PERFRINGENS AND CLOSTRIDIUM SEPTICUM TOXOID TO

CONTROL CELLULITIS IN TURKEYS

63

Summary

Clostridium perfringens and Clostridium septicum were recognized as the

causative agents for cellulitis in turkeys. The objective of this study was to develop and to

evaluate the use of a bivalent Clostridium perfringens and Clostridium septicum toxoid to

control cellulitis in commercial Nicholas White turkeys. A bivalent Clostridium

perfringens and Clostridium septicum toxoid was prepared and tested in six-week-old

commercial Nicholas White turkey poults under laboratory conditions for its safety and

efficacy. Vaccinated birds were then exposed to a homologous challenge. The mortality

and lesion development in vaccinated and unvaccinated controls were recorded and

compared. Blood samples from birds in both groups were examined to detect antibody

response to C. perfringens and C. septicum toxoid using ELISA. All the birds challenged

with C. perfringens and C. septicum in non-vaccinated group developed severe cellulitis

lesions and died within 24 h of challenge. There were several areas in the subcutaneous

tissues and muscles that were infiltrated by large numbers of inflammatory cells.

Numerous thick, rod-shaped bacteria were also present in the areas of inflammation in the

subcutaneous tissues and muscle fibers indicating the multiplication of bacteria. The

bivalent Clostridium perfringens and Clostridium septicum toxoid developed was found

to be completely safe and protective against homologous challenge. It produced

antibodies which appeared protective. Two doses of vaccine was found to be more

effective in protection than a single dose of the same preparation. The bivalent

Clostridium perfringens and Clostridium septicum toxoid could now be tested in turkeys

in the field to reduce losses due to cellulitis in turkeys.

64

INTRODUCTION

The prevalence and severity of cellulitis has increased over the last several years,

since the time it was first reported in 1939. This could be due to increasing regulations in

the use of in-feed antibiotic growth promoters (AGP) and restriction in antibiotic usage

(Li et al., 2010b). Currently, cellulitis in turkeys has been diagnosed in Minnesota,

Wisconsin, Missouri, Virginia, and other turkey producing areas (Clark et al., 2010).

Turkey industry in Minnesota and elsewhere is in need of an effective vaccine to

control cellulitis in their production facilities. The objective of this study was to develop

a Clostridium perfringens and Clostridium septicum formalin-inactivated toxoid (FIT)

vaccine for cellulitis in turkeys.

Clostridial vaccines containing toxoids and killed bacteria have been successfully

used in humans and animals against various clostridial infections. Sheep challenged with

C. perfringens toxoid were found to be protected against gas gangrene caused by C.

perfringens (Boyd et al., 1972).

In this study, we have examined the effect of immunization of turkeys with an

experimental bivalent C. perfringens and C. septicum toxoid in an attempt to control

cellulitis in turkeys under laboratory and field conditions.

MATERIAL AND METHODS

Preparation of a bivalent Clostridium perfringens and Clostridium septicum toxoid

Based on our previous studies (Chapter 3) we selected C. perfringens (UMNCP 1)

and C. septicum (UMNCS 106) to make a bivalent toxoid against cellulitis in turkeys.

The stock cultures of these isolates were grown in DS sporulation media and BHI (Oxoid,

65

Ogdensburg, NY) respectively for 24 h. At the end of 24 h, aliquots from each culture

were tested in mice for the toxin content. The MLD50 in mice for C. perfringens and C.

septicum was found to be 2.12 mg and 0.024 mg toxin units respectively. The cultures

were then treated with 0.5 % formalin and incubated at 37 C for 18 h for inactivation of

the toxin. The complete inactivation of the anaculture was confirmed by subculturing a

sample on anaerobic sheep blood agar plates and TSC agar plates anaerobically. Failure

of any growth was taken as complete inactivation of bacteria. The inactivation of the

toxins was also confirmed by hemolysis assay using sheep RBC.

The bivalent toxoid was mixed with vaccine grade mineral oil Dakreol-6VR

(Pennsylvania Refining Company, Butler, PA) as adjuvant and Arlacel-A (Sigma

Chemicals) as an emulsifier. Arlacel-A was used at the rate of 10 % (vol/vol) in the oil

component of toxoid. Briefly, for every 1,000 ml of the experimental bivalent toxoid

vaccine, the aqueous component contained bacterin-toxoid of C. perfringens (500 ml)

and C. septicum (7.81 ml). The oil component contained mineral oil (472.97 ml) and

Arlacel-A (19.22 ml). A water-in-oil emulsion was prepared after slowly mixing the oil

component in a blender to which the aqueous component was added over a period of 5

min. The final toxoid preparation was adjusted to contain 2 MLD50 toxin units per ml

each of inactivated C. perfringens and C. septicum preparation.

Laboratory evaluation of bivalent toxoid in turkey poults.

The experimental bivalent toxoid was first tested in six-week-old turkey poults for

its safety and efficacy. The birds were inoculated subcutaneously with either of the two

66

doses: a 1ml dose containing inactivated 2 MLD50 toxin units or a 2ml dose containing

inactivated 4 MLD50 toxin units of the toxoid preparation.

Briefly, forty eight six-week-old commercial Nicholas White turkey poults

obtained from a source with no history of cellulitis were divided into four groups (groups

I to IV) of twelve birds each. Birds in Group I were inoculated with 1 ml dose and birds

in group II were inoculated with 2ml dose each of the toxoid preparation subcutaneously

at the wing web. The birds in group III and IV were treated as non-vaccinated controls.

All the birds were monitored twice daily for any adverse effects of the toxoid. Any birds

showing severe pain and distress were to be euthanized in a carbon dioxide chamber. On

fourteenth day post-vaccination, half of the birds from group I and II were given a

booster dose of corresponding experimental toxoid at the same dose as the first vaccine.

The blood from all the birds was collected at 0, 14 and 28 days post- one-vaccination and

on 7-days-post booster vaccination for serological examination.

After 28-days post-vaccination, six birds each from groups I and II which

received one vaccination and six birds from non-vaccinated controls (group III) were

challenged with subcutaneous inoculation of 1 ml each of C. perfringens and C. septicum

culture containing 1.4x108 and 4.8 x 107spores/ml respectively on the breast region.

Similarly, after 14-days post- booster vaccination, remaining six birds each from groups I

and II and six birds from non-vaccinated controls (group III) were challenged .The

challenge dose we used in this experiment was the one we optimized in our previous

studies. The blood from all the birds were examined for sero-conversion to the vaccine

using an ELISA test. All the birds at the end of the study were euthanized in a carbon

dioxide chamber as per AVMA guidelines for euthanasia (2007).

67

All animal experimental protocols were approved by the Institutional Animal

Care and Use Committee (IACUC), and the procedures were performed in accordance

with the requirements. The animals were reared at Research Animal Resources (RAR)

isolation facilities at the University of Minnesota (Saint Paul, MN).

Statistical Analysis: Statistical analysis was performed by SAS software (SAS Language

Version 9.2, SAS Institute Inc, Cary, NC). The mortality data following challenge was

analyzed using Fishers exact test. The serological data was analyzed using GLM

procedure for repeated measures analysis of variance where a P value of 0.05 was

considered as significant.

RESULTS

The use of the experimental bivalent toxoid was found to be safe and no adverse

effects were noticed in any of the birds at the site of inoculation with either

administration of a 1 ml or 2 ml dose. All the birds challenged with C. perfringens and C.

septicum in non-vaccinated group (group III) developed severe cellulitis lesions (Figure

1A- 1F) and died within 24 h of challenge (Table 1).

A single administration of experimental bivalent toxoid at a 2 ml dose per bird

protected all birds post-challenge. No mortality or cellulitis lesion development was

observed in birds twice vaccinated and challenged in the vaccinated groups. Birds once

vaccinated with 1ml dose per bird lost one bird due to cellulitis within 48 h post-

challenge.

The serum antibody titers (OD values) obtained for C. perfringens and C.

septicum ELISA are shown in Tables 2 and 3 respectively. A significant rise in C.

68

perfringens and C. septicum antibody titers were noticed in the vaccinated birds on 14

and 28 days post one-vaccination or two vaccinations with either doses of the vaccine.

DISCUSSION

Subcutaneous inoculation of C. perfringens and C. septicum cultures consistently

produced cellulitis lesions and mortality in naïve turkeys in this study as in our previous

studies. The cellulitis lesions produced with (Figure 7A-F) incoculation of C. perfringens

and C. septicum cultures (Figure 7A-F) were comparable with the lesions produced

before with either C. perfringens or C. septicum cultures separately.

The experimental bivalent C. perfringens and C. septicum toxoid which we

developed was found to be completely safe and effective in reducing cellulitis lesions and

mortality in turkeys under laboratory conditions. There are a number of reports on the use

of Clostridial vaccines. Clostridial vaccines against C. perfringens and C. septicum

containing toxoids and killed bacteria have been successfully used before against clostridial

infections in humans and animals (Boyd et al., 1972). Similarly, a multicomponent C.

perfringens, C. septicum and Pasteurella bacterin was reported to reduce mortality in

gangrenous dermatitis affected broiler chickens (Gerdon et al., 1973).

Serum ELISA results showed significant antibody response against C. perfringens

as well as C septicum in vaccinated birds. Antibody response against C. septicum

bacterin/toxoid preparation was reported before in immunized turkeys under

experimental conditions (Tellez et al., 2009). In our laboratory studies, the use of the

toxoid was found to protect against cellulitis in all the vaccinated birds when compared

69

with non-vaccinated birds. Better protection and high serum antibody titers were noticed

in birds administered either 2 doses of 1 ml vaccination or one dose of 2 ml vaccination.

Our results offer promise for the use of C. perfringens and C. septicum toxoid

successfully to mitigate cellulitis cases in turkeys in the field.

In summary, the experimental bivalent C. perfringens and C. septicum toxoid we

developed offered a complete protection against cellulitis following homologous challenge

under experimental conditions. Use of a use of a higher concentration (2ml dose) of the

toxoid or a booster vaccination may offer a better protection against cellulitis due to C.

perfringens and C. septicum in turkeys.

70

TABLES

Table 1: Mortality following challenge with C. perfringens and C. septicum in

vaccinated and non-vaccinated birds (n=6).

Non-vaccinated Vaccinated

@ 1ml dose @ 2ml dose

Once vaccinated 6/6a 1/6a 0/6b

Twice vaccinated 6/6a 0/6a 0/6b

a,b Fraction of birds within a row with no common superscript are considered

significantly different (P<0.001).

71

Table 2: C. perfringens alpha-toxin ELISA serum antibody titer (OD values) following

vaccination with C. perfringens and C. septicum toxoid in vaccinated and non-vaccinated

birds.

Post-one-vaccination Post-two-vaccinations

OD Values OD Values

GROUPS 0dpv 14dpv 28dpv 7dp2v

Vac @ 1ml dose 0.3296 (0.07)a * 0.3597(0.06)b 0.4682(0.13)b 0.5963(0.13)c

Vac @ 2ml dose 0.3307 (0.06)a 0.34984 (0.05)b 0.5474(0.10)c 0.6006(0.20)c

Non-vaccinated 0.3155 (0.06)a 0.3386 (0.08)a 0.3212(0.08)a 0.3404 (0.07a)

* Values in paranthesis indicate standard deviation

dpv = days-post-vaccination.

abcd Means within a row with no common superscript are considered significantly

different (P < 0.05). 0.4682(0.13)b

72

Table 3: C. septicum ELISA serum antibody titer (OD values) following vaccination

with C. perfringens and C. septicum toxoid in vaccinated and non-vaccinated birds.


OD Values OD Values

0dpv 14dpv 28dpv 7dp2v

Vac @ 1ml dose 0.2701(0.07)a * 0.3309 (0.06)a 0.4095(0.12)b 0.5658(0.19)c

Vac @ 2ml dose 0.2880 (0.08)a 0.3441 (0.07)a 0.4984(0.16)b 0.644(0.15)c

Non-vaccinated 0.2889 (0.07)a 0.2465 (0.08)a 0.2884(0.07)a 0.2689(0.09)a

* Values in paranthesis indicate standard deviation

dpv = days-post-vaccination.


different (P < 0.05).

73

FIGURES

Figure 1. Gross lesions and histopathologic changes of subcutaneous and muscle tissue

in cellulitis in turkeys from C. perfringens and C. septicum challenge. Sections were

stained with H & E. (A) Blisters and vesicles extending from the breast region to

abdomen and tail regions in a bird inoculated with C. perfringens and C. septicum spore

culture. (B) The presence of froathy sanguinous subcutaneous exudate in the

subcutaneous tissue in a bird inoculated with C. perfringens and C. septicum spore

culture. (C) Histopathology section showing presence of moderate number of thick rod-

shaped bacteria in the subcutaneous tissue in a bird inoculated with C. perfringens and C.

septicum spore culture. Bar = 100 µm (D) Histopathology section showing presence of

multiple patches of muscle tissue in which the fibers were fragmented giving the areas a

“moth eaten” appearance in a bird inoculated with C. perfringens and C. septicum spore

culture Bar = 50 µm. (E) A nonvaccinated bird showing cellulitis lesions on the breast

region post-inoculation of C. perfringens and C. septicum. (F) A vaccinated bird showing

absence of any lesions on the breast region at 5-day-post inoculation of C. perfringens and

C. septicum toxoid.

74

75

CHAPTER VI

FIELD TRIALS WITH BIVALENT CLOSTRIDIUM PERFRINGENS AND

CLOSTRIDIUM SEPTICUM TOXOID

76

Summary

Cellulitis has emerged as a major problem in the turkey industry over the last few

years. Clostridium perfringens and Clostridium septicum were identified as the causative

agents for cellulitis in turkeys. The objective of this study was to field test the bivalent

Clostridium perfringens and Clostridium septicum toxoid we developed to reduce

cellulitis mortality and antibiotic usage in commercial turkeys. The bivalent Clostridium

perfringens and Clostridium septicum toxoid was evaluated for its use in the field in a

flock of commercial turkeys consisting of 16,000 birds where 8000 birds were vaccinated

and an equal number were kept as non-vaccinated controls. The birds were vaccinated at

six-weeks of age. Cellulitis related mortality in both groups was recorded and compared.

Blood samples from birds in both groups were examined to detect antibody response to

C. perfringens and C. septicum toxoid using ELISA. The use of bivalent Clostridium

perfringens and Clostridium septicum toxoid developed was found to be safe and

effective. It produced antibodies which appeared protective and did significantly reduce

antibiotic usage to control cellulitis as well as mortality. Though the toxoid did not

offered complete protection in the field the use of a booster dose may help in better

protection against cellulitis in turkeys.

77

INTRODUCTION

With permission from the Board of animal health and University of Minnesota

and consent from a leading commercial turkey producer in Minnesota, a field trial was

conducted in a cellulitis endemic farm to test the efficacy of our experimental bivalent C.

perfringens and C. septicum toxoid.


Experimental design: The commercial turkey brooder barn we selected typically broods

16,000 turkey poults. All the birds were male poults that were vaccinated for Newcastle

(3d & 3wk) and hemorrhagic enteritis (HE) (4wk). One half of the birds (n=8,000) at the

age of 6-weeks were selected randomly from a single brooder barn for vaccination. One

milliliter dose of the vaccine was administered subcutaneously at the neck region using a

vaccine applicator.

Following vaccination, all the vaccinated and non-vaccinated groups were

transferred to grower farms where there was a consistent history of cellulitis. The

vaccinated and non-vaccinated birds were kept separated in the same barn by putting a

firm wire-mesh separation in the middle of each barn. The birds were likely to be

exposed to Clostridia found in the poultry environment (natural challenge). A therapeutic

antibiotic penicillin G sodium usage in water was in place as a precautionary measure as

and when mortality rose above 0.1% per day. Daily mortality and packs of penicillin (one

pack of penicillin contains 1.0 billion I.U. (601.2 g) of Penicillin G Potassium) used to

control mortality was recorded in both vaccinated and non-vaccinated birds from 13 to

22 weeks of age. In the farm, there is routine monitoring of all sick, moribund and dead

78

birds twice daily and those identified sick and moribund were euthanized. The results

were compared to detect any observable differences in the usage of antibiotics or

occurrence of cellulitis related mortality between vaccinated and non-vaccinated birds.

Following field vaccination, five birds each at 10-weeks of age from vaccinated and

non-vaccinated groups from each farms were transferred to RAR facilities at University of

Minnesota. Blood was collected for ELISA and all the birds were challenged as mentioned

before. The cellulitis development and mortality between the groups was recorded. All the

birds at the end of the study were euthanized in a carbon dioxide chamber.

Statistical analysis: Statistical analysis was performed using SAS software (SAS

Language Version 9.2, SAS Institute Inc, Cary, NC). The mortality data following

challenge with spore culture was analyzed using Fishers exact test. A survival analysis of

maximum likelihood estimates using proportional hazards model (PH-REG procedure)

was conducted to analyze the mortality rates in vaccinated and non-vaccinated groups. A

Pearson’s Chi square test was used to compare the use of antibiotics between vaccinated

and non-vaccinated groups. The serological data was analyzed using GLM procedure for

repeated measures analysis of variance where a P value of 0.05 was considered as

significant.

79

RESULTS

Field trials with bivalent toxoid in commercial turkeys

Mortality in non-vaccinated and vaccinated birds are shown in Table 1 and Figure

1. There were significant differences in the mortality (P<0.001) between non-vaccinated

group and vaccinated group. The use of this experimental bivalent toxoid reduced

mortality from 9.4 % to 7.4 %. This accounts to 21% reduction in cellulitis related

mortality in vaccinated birds when compared to non-vaccinated birds. A Kaplan Meier

survival curve comparing the survival distribution function of non-vaccinated and

vaccinated birds was shown in Figure 2. The Hazard ratio was found to be 1.3 for the

non-vaccinated group over vaccinated group of birds.

The need for use of antibiotic penicillin was significantly reduced because of

vaccination (P<0.0001) from 547 packs in non-vaccinated birds to 361 packs in vaccinated

birds (Figure 3). This accounts for a 35% reduction in the use of antibiotics in vaccinated

birds when compared with non-vaccinated birds.

Laboratory challenge studies with bivalent C. perfringens and C. septicum

vaccinated and non vaccinated birds from the field.

When we challenged ten birds transferred from vaccinated and non-vaccinated

groups from field trials at the university, we observed a mortality of 80% (8/10) in 24 h in

non-vaccinated birds at 95% confidence level with a confidence interval of 55-100%. The

two birds which survived did show signs of cellulitis lesions at 2-6 days but the lesions

subsided in few days. All birds from vaccinated group resisted challenge and did not

develop any signs of cellulitis lesions or mortality and were found to be completely

80

protected. Vaccinated birds had C. perfringens alpha toxin ELISA serum antibody titer of

0.4861 +/- 0.09 and non-vaccinated birds had an antibody titer of 0.3945 +/- 0.10. The C.

septicum ELISA serum antibody titers were 0.4582 +/- 0.11 and 0.3654 +/- 0.13

respectively.

DISCUSSION

In our laboratory studies, better protection were seen in birds administered either

2 doses of 1 ml vaccination or one dose of 2 ml vaccination. However, administration of

a volume of greater than 1 ml was not tried under field conditions due to difficulties in

labor and handling.

When a single administration at 1ml dose of our experimental bivalent C.

perfringens and C. septicum toxoid was used under field conditions, it was found to be

protective and it reduced number of cellulitis cases, antibiotic usage and mortality in

commercial turkeys. The toxoid significantly reduced the mortality by 21 % in vaccinated

birds when compared with non-vaccinated birds. A 21 % reduction in mortality is highly

significant, considering the fact that the disease appears in adult birds and the normal

mortality rate at this age group is negligible in commercial settings. The reason for a

partial protection was assumed to be due to insufficient antigen concentration in a single

dose of the vaccine. Being a toxoid, multiple vaccinations may be required for prolonged

protection.

The use of C. septicum bacterin/toxoid preparation elicited antibody response

against C. septicum in immunized turkeys under experimental conditions. Tellez et al.,

81

2009 also reported similar findings in vaccinated turkeys however no challenge studies

were conducted to demonstrate protection.

The incidence of C. perfringens and C. septicum associated diseases in poultry has

increased significantly in the recent years because of the reduced use of antimicrobial

growth promoters (van Immerseel et al., 2004). Cellulitis is commonly controlled in a

preventative manner now by incorporation of antimicrobial drugs in the feed or water, but

this practice is increasingly criticized or has been banned in some countries. Since cellulitis

cases in turkeys are also rising at an alarming rate, there is a need to investigate alternative

biologic approaches for its effective control. We were able to reduce the dose and duration

of antibiotic Penicillin G sodium usage significantly in turkeys where the vaccine was used

to control cellulitis and mortality when compared with non-vaccinated controls.

It is widely accepted that the natural outbreaks of cellulitis in turkeys are associated

with proliferation of pathogens in the poultry environment. Under field conditions, the

commercial poults are reared on deep litter systems and the litter is not replaced for every

flocks. Build up of Clostridial load may occur during this time. A greater environmental

load may help these Clostridial spores to persist and also may increase the chance of

fecal-oral route of spread over time. Under such scenarios, adverse effects of greater

severity may be expected in the field when birds were exposed to extremely high dose of

Clostridial challenge than under experimental conditions. However additional studies are

required to know the effect of Clostridial load in the litter on the incidence of cellulitis

cases under field and experimental conditions.

With single vaccination in two farms, in antibiotic cost alone the industry saved

approximately 4278 dollars considering the fact that each pack of penicillin costs 23

82

dollars. When reduction in mortality is also taken into account, the average savings per

barn could be more than six thousand dollars. Our results establish the fact that C.

perfringens and C. septicum toxoids could be used successfully to mitigate cellulitis cases

in turkeys in the field.

In summary, the experimental bivalent C. perfringens and C. septicum toxoid we

developed offered a complete protection against cellulitis following homologous challenge

under experimental conditions. The same vaccine enabled us to reduce the mortality and

use of antibiotic treatment in preventing cellulitis in commercial turkeys. Multiple

vaccinations as well as use of a higher concentration of antigens or a more refined toxoid

may offer a better protection against cellulitis due to C. perfringens and C. septicum in

turkeys.

83

TABLES

Table 1. Comparison of mortality and antibiotic usage post-vaccination with C.

perfringens and C. septicum toxoid in non-vaccinated and vaccinated birds from 13-weeks

to 22-weeks of age.

Non-vaccinated group Vaccinated group

Number of birds 8,000 8,000

Mortality 748 595 a

% Mortality 9.4 7.4 a

Total penicillin usage* 547 packs 361 packs a

Penicillin usage days 59 31a

a indicates P < 0.05

* one pack of penicillin contains 1.0 billion I.U. (601.2 g) Penicillin G Potassium

84

Figure 1. Mortality following vaccination with a bivalent C perfringens and C. septicum

toxoid in non-vaccinated (line) and vaccinated birds (broken line) from 13-weeks of age

until 22-weeks of age.

0

5

10

15

20

25

30

35

92 95 98 101 104 107 110 113 116 119 122 125 128 131 134 137 140 143 146 149 152 155

Age in days

Dai

ly M

orta

lity

Non Vac

Vac

85

Figure 2: Kaplan Meier survival curve showing the survival distribution function

following vaccination with bivalent C perfringens and C. septicum toxoid in non-

vaccinated (broken line) and vaccinated birds (line) from 13-weeks of age until 22-weeks

of age.

0.9

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1

90 100 110 120 130 140 150 160Age in days

surv

ival

dis

t. fu

nctio

n

Non Vac

Vac

86

Figure 3. Antibiotic usage following vaccination with bivalent C perfringens and C.

septicum toxoid in non-vaccinated (black bars) and vaccinated birds (white bars) 13-

weeks of age until 22-weeks of age.

0

5

10

15

20

25

94 97 100

103

106

109

112

115

118

121

124

127

130

133

136

139

142

145

148

151

154

Age in days

Pe

nic

illin

pa

cks Non Vaccinated

Vaccinated

87

CHAPTER VII

LABORATORY STUDIES ON CLOSTRIDIUM SEPTICUM TOXOID TO

CONTROL CELLULITIS IN TURKEYS

88

Summary

Isolation results from cellulitis cases from various diagnostic laboratories across

the state indicated an increased isolation rate of Clostridium septicum alone from

cellulitis cases in the recent years. More over, results from our study also suggested

increased potency for Clostridium septicum to cause cellulitis in turkeys than Clostridium

perfringens. The objective of this study was to develop and to evaluate the use of a

Clostridium septicum anaculture toxoid alone to control cellulitis in commercial turkeys.

A Clostridium septicum toxoid was prepared and tested in six-week-old commercial

Nicholas White turkey poults under laboratory conditions for its safety and efficacy.

Vaccinated birds were exposed to a homologous challenge. The mortality and lesion

development in vaccinated and unvaccinated controls were recorded and compared.

Blood samples from birds in both groups were examined to detect antibody response to

C. septicum toxoid using ELISA. The birds challenged with C. septicum in vaccinated

group did not show any cellulitis lesions or mortality where as birds in the non-

vaccinated group (group III) developed severe cellulitis lesions and died within 24 h of

challenge. Better protection were seen in birds administered either 2 doses of 1ml

vaccination or one dose of 2ml vaccination. The Clostridium septicum toxoid developed

was found to be completely safe and effective against homologous challenge. It produced

antibodies which appeared protective. The Clostridium septicum toxoid developed could

now be tested in the field to reduce losses due to cellulitis in turkeys.

89

INTRODUCTION

Over the years we noticed a change in the isolation pattern of Clostridial isolates

from cellulitis cases in turkeys. Data from diagnostic labs as wells as from turkey field

veterinarians (personel communication) indicated an increased isolation of C. septicum as

against C. perfringens from cellulitis cases. Our experimental data showed greater

potency for C. septicum than C. perfringens to cause cellulitis in turkeys. Recent studies

(Tellez et al., 2009) have drawn attention to C. septicum and their role in causing cellulitis.

The objective of this study was to develop and test a Clostridium septicum alone

toxoid in the changing circumstances for cellulitis in turkeys. In this study, we have

examined the effect of immunization of turkeys with an experimental C. septicum toxoid

vaccine in an attempt to control cellulitis in turkeys under laboratory and commercial

conditions.


Bacterial isolates and growth conditions: Based on our previous studies with

Clostridial isolates for their spore and toxin production and mice lethal assay, we selected

C. septicum (UMNCS 106) isolate for vaccine production. It was grown in BHI at 37 C

for 18 h and 25 C for 6 h. The cultures were then treated with 0.5 % formalin and

incubated at 37 C for 18 h for inactivation of the toxin. The complete inactivation was

confirmed by subculturing a sample on anaerobic sheep blood agar plates anaerobically.

Failure of any growth was taken as complete inactivation of bacteria. The inactivation of

the toxins was also confirmed by hemolysis assay using sheep RBC.

90

Preparation of experimental formalin-inactivated toxoid (FIT)

An experimental toxoid was prepared using formalin-inactivated C. septicum

culture and a vaccine grade mineral oil Dakreol-6VR as adjuvant and Arlacel-A as an

emulsifier. Arlacel-A was used at the rate of 10 % (vol/vol) in the oil component of

toxoid. Briefly, for every 1,000 ml of the experimental bivalent toxoid vaccine, the

aqueous component contained bacterin-toxoid of C. septicum (500 ml). The oil

component contained mineral oil (450 ml) and Arlacel-A (50 ml). A water-in-oil

emulsion was prepared after slowly mixing the oil component in a blender to which the

aqueous component was added over a period of 5 minutes. The final toxoid preparation

was adjusted to contain 128 MLD50 toxin units per ml of inactivated C. septicum

preparation. The experimental birds were then vaccinated at different doses to evaluate

the efficacy of the vaccine at each dose rate. The different dose rates examined are 1ml

(128 MLD 50 toxin units) and 2ml (256 MLD 50 toxin units).

Laboratory evaluation of C. septicum toxoid in turkey poults.

The experimental bivalent toxoid was first tested in six-week-old turkey poults for

its safety and efficacy. The birds were inoculated subcutaneously with either of the two

doses: a 1 ml dose containing inactivated 128 MLD50 toxin units or a 2 ml dose

containing inactivated 256 MLD50 toxin units of the toxoid preparation.

Briefly, ninety six, six-week-old commercial Nicholas White turkey poults

obtained from a source with no history of cellulitis were divided into four groups (groups

I to IV) of twenty four birds each. Birds in Group I were inoculated with 1ml dose and

birds in group II were inoculated with 2ml dose each of the toxoid preparation

91

subcutaneously at the wing web. The birds in group III and IV were treated as non-

vaccinated controls. All the birds were monitored twice daily to see for any adverse

effects of the toxoid. Any birds showing severe pain and distress were planned to be

euthanized in a carbon dioxide chamber. On fourteenth day post-vaccination, half of the

birds from group I and II were given a booster dose of corresponding experimental toxoid

at the same dose as the first vaccine. The blood from all the birds was collected at 0, 14

and 28 days post- one-vaccination and on 7-days-post booster vaccination for serological

examination.

After 28-days post-vaccination, twelve birds each from groups I and II which

received one vaccination and twelve birds from non-vaccinated controls (group III) were

challenged with subcutaneous inoculation of 2 ml of C. septicum culture containing 4.8 x

107spores/ml respectively on the breast region. Similarly, after 14-days post- booster

vaccination, remaining twelve birds each from groups I and II and six birds from non-

vaccinated controls (group III) were challenged. The sera from all the birds were

examined for sero-conversion to the vaccine using an ELISA test. All the birds at the end

of the study were euthanized in a carbon dioxide chamber as per AVMA guidelines for

euthanasia (2007).

All animal experimental protocols were approved by the Institutional Animal

Care and Use Committee (IACUC), and the procedures were performed in accordance

with the requirements. The animals were reared at Research Animal Resources (RAR)

isolation facilities at the University of Minnesota (Saint Paul, MN).

92

RESULTS

Laboratory evaluation of C. septicum toxoid in turkey poults.

No adverse effects were noticed in any of the birds at the site of inoculation of

experimental bivalent toxoid with either administration of 1ml or 2ml dose of the toxoid.

All the birds challenged with C. septicum in non-vaccinated group (group III) developed

severe cellulitis lesions and died within 24 h of challenge (Table 1).

Groups given 1ml dose rate (128 MLD 50 units) per bird lost one bird with in 48

h of inoculation. Two vaccinations with a low dose (1 ml) of toxoid were as protective as

one or two doses of high dose (2 ml) of toxoid.

A single administration of experimental bivalent toxoid at a 2ml dose per bird

protected all birds post-challenge. No mortality or cellulitis lesion development was

observed in birds twice vaccinated and challenged in the vaccinated groups. Birds once

vaccinated with 1 ml dose per bird lost one bird due to cellulitis within 48 h post-

challenge.

The serum antibody titer (OD values) obtained by ELISA is shown in Table 2. A

significant rise in C. septicum antibody titers were noticed in the vaccinated birds on 14

and 28 days-post-one-vaccination or two with either doses of the vaccine.

DISCUSSION

Subcutaneous inoculation of C. septicum cultures consistently produced cellulitis

lesions and mortality in naïve turkeys in this study as in our previous studies. Similar

descriptions of lesions caused by C. perfringens (Carr et al., 1996) have been reported

before but not with C. septicum in turkeys.

93

Serum ELISA results showed significant antibody response against C perfringens

as well as C septicum in vaccinated birds. Antibody response against C. septicum

bacterin/toxoid preparation was reported before in immunized turkeys under

experimental conditions (Tellez et al., 2009). The experimental C. septicum toxoid which

we developed was found to be completely safe and effective in reducing cellulitis lesions

and mortality in turkeys under laboratory conditions. Better protection and higher serum

antibody titers were seen in birds administered either 2 doses of 1 ml vaccination or one

dose of 2 ml vaccination. Our results offer promise for the use of C. septicum toxoid

successfully to mitigate cellulitis cases in turkeys in the field.

In summary, the experimental C. septicum toxoid we developed offered complete


conditions. Use of a higher concentration (2ml dose) of the toxoid or a booster

vaccination may offer better protection against cellulitis due to C. septicum in turkeys.

94

TABLES

Table 1: Mortality following challenge with C. septicum in C. septicum vaccinated and

non-vaccinated birds (n=12).

Non-vaccinated Vaccinated

@ 1 ml dose @ 2 ml dose

Once vaccinated 12/12b 1/12a 0/12a

Twice vaccinated 12/12a 0/12a 0/12a

a,b Fraction of birds within a row with no common superscript are considered

significantly different (P<0.001).

95

Table 2: C. septicum ELISA serum antibody titer (OD values) following vaccination with

C. septicum toxoid in vaccinated and non-vaccinated birds (n=12).


OD Values OD Values

0dpv 14dpv 28dpv 7dp2v

Vac @ 1 ml dose 0.2459 (0.13)*a 0.6574 (0.24)b 0.704(0.10)b 0.9217(015)c

Vac @ 2 ml dose 0.2691(0.09)a 0.7645 (0.23)b 0.7254(0.09)b 1.0686(0.14)c

Non-vaccinated 0.2708 (0.54)a 0.2182 (0.17)a 0.2814(0.05)a 0.2689(0.05)a

* values in paranthesis indicate standard deviation; dpv = days-post-vaccination.


different (P<0.05).

96

CHAPTER VIII

FIELD TRIAL WITH CLOSTRIDIUM SEPTICUM TOXOID

97

Summary

Cellulitis continued to be a major problem in the turkey industry in the recent

years irrespective of all the control measures adopted in the field. Cellulitis due to

Clostridium septicum in turkeys increased at an alarming rate. The objective of this study

was to field test the C. septicum toxoid we developed to reduce cellulitis mortality and

antibiotic usage in commercial Nicholas White turkeys. The Clostridium septicum toxoid

was evaluated for its use in the field in a flock of commercial turkeys consisting of

16,000 birds where 8000 birds were vaccinated and an equal number were kept as

unvaccinated controls. The birds were vaccinated at six-weeks of age. The mortality in

both vaccinated and unvaccinated groups was recorded and compared. Blood samples

from birds in both groups were examined to detect antibody response to C. septicum

toxoid using ELISA. In another field trial, we vaccinated 2000 birds in a barn consisting

of 8000 birds and both groups were kept together in same barns. Cellulitis mortality was

recorded in vaccinated and non-vaccinated birds. The use of Clostridium septicum toxoid

developed was found to be safe and effective in both trials. Vaccination produced

antibodies and did significantly reduce antibiotic usage to control cellulitis as well as

mortality. Though a single vaccination with C. septicum toxoid did not offered complete

protection in the field the use of multiple vaccinations may elicit a better antibody

response and protection against cellulitis in turkeys. The results of our study offer

promise in reducing antibiotic usage and also to reduce mortality due to cellulitis in

turkeys due to C. septicum.

98

INTRODUCTION

With permission from the Board of animal health and University of Minnesota

and consent from a leading commercial turkey producer, field trials were planned to test

the efficacy of our experimental bivalent toxoid. We conducted two field studies under

two commercial settings.


Experimental design for Study I: The experimental design in this study was similar to

the previous vaccination trial explained in Chapter IV. Briefly, Four thousand poults at

the age of 6-weeks were vaccinated in a flock of 8,000 Nicholas White turkey poults in a

brooder barn with experimental bivalent toxoid with 1x dose. All the birds were male

poults that have been vaccinated for Newcastle (3d & 3wk) and hemorrhagic enteritis

(HE) (4wk). One half of the birds were selected randomly from a single brooder barn for

vaccination. The vaccine was administered subcutaneously at the neck region using a

vaccine applicator.

Following vaccination, 2000 birds each from vaccinated and non-vaccinated

group were transferred to two different grower Farms (Farm I and Farm II) where there

was a consistent history of cellulitis. The vaccinated and non-vaccinated birds were

separated by a wire-mesh and subjected to similar management conditions. The birds

were likely to be exposed to Clostridia found in the poultry environment (natural

challenge). A therapeutic antibiotic penicillin G sodium usage in water was in place as a

precautionary measure as and when cellulitis related mortality rose above 0.1% per day.

Daily mortality and packs of penicillin (one pack of penicillin contains 1.0 billion I.U.

99

(601.2 g) of Penicillin G Potassium) used to control cellulitis related mortality was

recorded in both vaccinated and non-vaccinated birds from 10-weeks of age until 22-

weeks of age. In the farm, there is routine monitoring of all the sick, moribund birds and

dead birds twice daily and those identified sick and moribund were euthanized. The

results were compared to detect any observable differences in the occurrence of mortality

between vaccinated and non-vaccinated birds.

Ten birds at 10-weeks of age from vaccinated and non-vaccinated groups from the

field vaccinated trials were transferred to RAR facilities at University of Minnesota. All the

birds were challenged as mentioned before. The cellulitis development and mortality

between the groups was recorded. All the birds at the end of the study were euthanized in

a carbon dioxide chamber.

Statistical analysis was performed by SAS software (version 9.2)e. The mortality

data following challenge with spore culture was analyzed using Fishers exact test. A

survival analysis of maximum likelihood estimates using proportional hazards model

(PH-REG procedure) was conducted to analyze the mortality rates in vaccinated and non-

vaccinated groups. A Pearson’s Chi square test was used to compare the use of antibiotics

between vaccinated and non-vaccinated groups. The serological data was analyzed using

GLM procedure for repeated measures analysis of variance where a P value of 0.05 was

considered as significant.

Experimental design for Study II: In the second experimental model, the vaccinated

birds and non vaccinated birds were kept together to eliminate any bias towards exposure

to Clostridia from the environment with in the barn. Briefly, 4,000 poults at the age of 6-

100

weeks were vaccinated in a flock of 20,000 Nicholas White turkey poults in a brooder

barn with experimental bivalent toxoid with 1x dose. All the birds were male poults that

have been vaccinated for Newcastle (3d & 3wk) and hemorrhagic enteritis (HE) (4wk).

The vaccinated birds used were toe clipped for identification purpose. The vaccine was

administered subcutaneously at the neck region using a vaccine applicator.

Following vaccination, 2000 birds each from vaccinated and 8000 birds each from

non-vaccinated group were transferred to two different grower Farms (Farm I and Farm

II) where there was a consistent history of cellulitis. The vaccinated and non-vaccinated

birds were kept together and subjected to similar management conditions. The birds were

likely to be exposed to Clostridia found in the poultry environment (natural challenge). A

therapeutic antibiotic penicillin G sodium usage in water was in place as a precautionary

measure as and when the mortality due to cellulitis rose above 0.1% per day. Daily

mortality with visible cellulitis lesions were recorded in both vaccinated and non-

vaccinated birds until the birds were marketed. In the farm, there was routine monitoring

of all the sick, moribund birds and dead birds twice daily and those identified sick and

moribund were euthanized. The results were compared to detect any observable

differences in the occurrence of mortality between vaccinated and non-vaccinated birds.

RESULTS

The mortality due to cellulitis in vaccinated and non-vaccinated birds are shown in

Table 1 and Figure 1. There were significant differences in the mortality (P<0.005) in the

vaccinated group from non vaccinated groups. The use of experimental C. septicum

toxoid reduced cellulitis-related mortality from 12.1 % to 10.5 %. This accounts for a 14%

101

reduction in cellulitis-related mortality in vaccinated birds when compared with non-

vaccinated birds.The Hazard ratio was found to be 1.14 for the non-vaccinated group

over vaccinated group of birds. A Kaplan Meier survival curve comparing the survival

distribution function of non-vaccinated and vaccinated birds was shown in Figure 2. No

significant differences was noticed in terms of mortality between the farms (P=0.6343).

The need for use of antibiotic penicillin was significantly reduced because of

vaccination (P<0.001) from 204 packs in non-vaccinated birds to 84 packs in vaccinated

birds. This accounts for a 59% reduction in the use of antibiotics in vaccinated birds when

compared with non-vaccinated birds. The days required to use peniciilin to control

cellulitis was also reduced from 68 days in non-vaccinated group to to 28 days in

vaccinated group (Figure 3).

In the second study, C. septicum toxoid significantly reduced cellulitis mortality in

vaccinated birds from 1.68 % to 0.87 %, which accounts approximately 50% reduction in

mortality (Table 2).

Challenge studies with experimental formalin-inactivated toxoid (FIT) vaccinated

birds against Clostridium septicum.

Ten birds each from vaccinated and nonvaccinated groups from the field vaccinated

trials were transferred to RAR facilities at University of Minnesota. The blood was

collected at 3 weeks post vaccination from the vaccinated and the nonvaccinated birds and

all the birds were challenged. Challenge inoculum and dose used was the same as in our

previous studies. We observed a mortality of 100% (10/10) in 24 h in non-vaccinated birds.

All birds from vaccinated group resisted challenge and did not develop any signs of

102

cellulitis lesions or mortality and were found to be completely protected. Vaccinated birds

had C. septicum ELISA serum antibody titer of 0.6870 +/- 0.12 and non-vaccinated birds

had an antibody titer of 0.3626 +/- 0.09.

DISCUSSION

When a single administration at 1ml dose of our experimental bivalent or C.

septicum toxoid was used under field conditions, it was found to be protective and it did

reduce the number of cellulitis cases, antibiotic usage and mortality due to cellulitis in

commercial turkeys.

The toxoid significantly reduced the mortality by 14% in vaccinated birds when

compared with non-vaccinated birds. A 14% reduction in mortality is highly significant,

considering the fact that the disease appears in adult birds and the normal mortality rate at

this age group is negligible in commercial settings. The reason for a partial protection

was assumed to be due to single vaccination or an insufficient antigen concentration in a

single dose of the vaccine.

The use of C. septicum bacterin/toxoid preparation was reported to elicit antibody

response against C. septicum in immunized turkeys under experimental conditions

(Tellez et al., 2009). Our results establish the fact that C. septicum toxoids could be used

successfully to mitigate cellulitis cases in turkeys in the field. However, being a toxoid

multiple vaccinations are required to boost the immunity for longer periods until the birds

are marketed. The use of a formalin-inactivated C. difficile toxoid in humans (Kotloff et

al., 2001) as well as use of a formalin inactivated C. septicum alpha toxoid in mice

(Amimoto et al., 2002) were found to be protective but the immunity was short lived. In

103

another study, a multicomponent C. perfringens, C. septicum and Pasteurella bacterin was

reported to reduce mortality in gangrenous dermatitis affected broiler chickens however

no information is available about the duration of immunity (Gerdon, 1973).

With single vaccination in two farms, in antibiotic cost alone the industry saved

approximately 2760 dollars considering the fact that each pack of penicillin costs 23

dollars. When reduction in mortality is also taken into account, the average savings per

barn could be around five thousand dollars.

In summary, the experimental C. septicum toxoid we developed offered complete


conditions. The same vaccine was found to be effective in reducing mortality and use of

antibiotic treatment in preventing cellulitis in commercial turkeys. In the future, use of

multiple vaccinations as well as use of day-old-vaccination has to be considered for

reducing further losses due to cellulitis in commercial turkey production.

104

TABLES

Table 1. Comparison of cellulitis related mortality and antibiotic usage post-vaccination

with C. septicum toxoid in non-vaccinated and vaccinated birds from 10-weeks of age to

22-weeks of age.



Mortality 970 840 a

% Mortality 12.1 10.5 a

Penicillin usage * 204 packs 84 packs a

Penicillin usage days 68 28 a


* one pack of penicillin contains 1.0 billion I.U. (601.2 g) Penicillin G Potassium

105

Table 2. Comparison of cellulitis related mortality in C. septicum toxoid vaccinated and

non-vaccinated birds from 7-weeks to 19-weeks of age.



Cellulitis Mortality 268 35 a

% Cellulitis Mortality 1.68 % 0.87 % a


106

FIGURES

Figure 1. Mortality following vaccination with C. septicum toxoid in non-vaccinated

(line) and vaccinated birds (broken line) from 10-weeks of age until 22-weeks of age.

0

5

10

15

20

25

30

35

40

45

50

Age in days

Daily

Mortalit

y

Vac

Non Vac

107

Figure 2. Kaplan Meier survival curve showing the survival distribution function

following vaccination with C. septicum toxoid in non-vaccinated (broken line) and

vaccinated birds (line) from 10-weeks of age until 22-weeks of age.

0.86

0.88

0.9

0.92

0.94

0.96

0.98

1

60 80 100 120 140 160Age in days

Sur

viva

l dis

t. fu

nctio

n

Non Vac

vac

108

Figure 3. Antibiotic usage following vaccination with C. septicum toxoid in non-

vaccinated (black bar) and vaccinated birds (white bar) from 10-weeks of age until 22-

weeks of age.

0

1

2

3

4

5

6

7

70

75

80

85

90

95

100

105

110

115

120

125

130

135

140

145

150

155

Age in days

Penic

illin

pack

s

Vac

Non Vac

109

CHAPTER IX

CONCLUSIONS

110

Cellulitis has continued to cause extensive losses in turkey production in USA due

to severe mortality, carcass condemnation and treatment costs. Clostridium perfringens and

Clostridium septicum are considered as the primary causative agents of cellulitis in turkeys.

The results of this study emphasized the role of these two agents in causing cellulitis in

turkeys. Cellulitis lesions and mortality in turkeys were successfully reproduced in

turkeys with either Clostridium perfringens or Clostridium septicum isolated from

cellulitis cases. Clostridium perfringens spore cultures produced mortality only at higher

doses whereas C. septicum caused mortality even at the lower dose in turkeys. This

necessitates the need to reduce exposure to both Clostridium perfringens and Clostridium

septicum to control cellulitis in commercial turkey production.

Clostridium perfringens and Clostridium septicum isolates were found to vary in

their abilities to produce spores as well as toxins. We observed differences in the toxicity

and biological effects of different strains of C. perfringens and C. septicum either in vitro,

or in vivo in mice as well as in turkeys. Though the spore count and hemolytic effects of

C. perfringens were found to be higher than C. septicum in vitro, the lethal effects in

mice as well as in turkeys were not comparable. Clostridium septicum was found to be

more potent than Clostridium perfringens in causing cellulitis in turkeys. Control

measures could now be targeted more on C. septicum to reduce losses in turkey

production. The development of fulminant cellulitis lesions and mortality was markedly

higher in seven-week-old birds than in three-week old birds. The phylogenetic tree

constructed from C. septicum isolates indicated a high level of conservation present

within the genome of C. septicum. These results will enable us in selecting a suitable

vaccine candidate for the control of cellulitis.

111

The major secretory toxins we identified in C. perfringens isolates from cellulitis

cases by MALDI-TOF mass spectrometry were phospholipase, collagenase,

hyaluronidase, Dnase, enolase, muramidase, pyruvate kinase and hypothetical proteins.

We observed two distinct proteomic profiles for C. perfringens isolates and the role of

hypothetical proteins in cellulitis need to be further investigated. We observed only one

type of proteomic profile for C. septicum isolates. The major secretory toxins we

identified in C. septicum were alpha toxin, septicolysin, sialidase, Dnase, flagellin and

hypothetical proteins.

The gross and microscopical lesions obtained after administration of spore

cultures of either C. perfringens or C. septicum in seven-week-old birds were comparable

to the cellulitis lesions reported before either with C. perfringens experimental infection or

field observations. However, multiplication of either C. perfringens or C. septicum in the

muscle tissues of cellulitis affected birds has not been reported before. C. septicum appears

to be more potent in causing severe cellulitis lesions and higher mortality in turkeys

although the role of C. perfringens in causing fatal cellulitis cannot be ignored. Our

cellulitis disease model offer promise to use it as a challenge model in the development

of vaccines against cellulitis in turkeys.

The experimental bivalent C. perfringens and C. septicum toxoid as well as C.

septicum toxoid we developed offered complete protection against cellulitis following

homologous challenge under experimental conditions. The use of these vaccines was found

to be effective in reducing mortality and use of antibiotic treatment due to cellulitis

significantly in commercial turkeys.

On future directions, an immediate follow up of the present study would be to

112

develop a day-old bivalent or a C. septicum toxoid. A vaccination protocol with a day-old

vaccination followed by a booster at six-weeks of age may elicit better protection against

cellulitis due to C. perfringens and C. septicum in turkeys. Multiple vaccinations as well

as use of a higher concentration of antigens need to be considered. This approach will

eventually help us to control cellulitis in turkey flocks of Minnesota.

113

CHAPTER X

REFERENCES

114

Al-Sheikhly F, and Truscott RB. The interaction of Clostridium perfringens and its toxins

in the production of necrotic enteritis of chikens. Avian Dis. 21(2):256-263,

1977.

Amimoto K, Ohgitani T, Sasaki O, Oishi E, Katayama S, Isogai M, and Ota S. Protective

effect of Clostridium septicum alpha-toxoid vaccine against challenge with

spores in guinea pigs. J Vet Med Sci. 64(1):67-69, 2002.

Ballard J, Bryant A, Stevens D, and Tweten RK. Purification and characterization of the

lethal toxin (alpha-toxin) of Clostridium septicum. Infect Immun. 60(3):784-790,

1992.

Boyd NA, Thomson RO, Walker PD. The preventions of experimental Clostridium novyi

and C. perfringens gas gangrene in high velocity missile wounds by active

immunization. J med microbio. 5:467-472, 1972.

Buchanan AG. Clinical laboratory evaluation of a reverse CAMP test for presumptive

identification of Clostridium perfringens. J Clin Microbiol. 16(4):761-762, 1982.

Carr D, Shaw D, Halvorson DA, Rings B, and Roepke D. Excessive mortality in market-

age turkeys associated with cellulitis. Avian Dis. 40:736-741, 1996.

Carr Debb, The etiology and pathogenesis of turkey cellulitis. MS thesis submitted to

University of Minnesota, 1996.

Chalmers G., Bruce HL, Toole DL, Barnum DA, and P. Boerlin. Necrotic Enteritis

Potential in a Model System Using Clostridium perfringens Isolated from Field

Outbreaks Avian Dis. 51(4):834–839, 2007.

Clark S, Porter R, McComb B, et al. Clostridial Dermatitis and Cellulitis: An Emerging

Disease of Turkeys. Avian Dis. 54:788-794, 2010.

115

Cortiñas TI, Mattar MA, and Stefanini de Guzmán AM. Alpha-toxin production by

Clostridium septicum at different culture conditions. Anaerobe. 3(2-3):199-202,

1997.

Craven SE, Cox NA, Stern NJ, and Mauldin JM. Prevalence of Clostridium perfringens

in commercial broiler hatcheries. Avian Dis. 45(4):1050-1053, 2001a.

Craven SE, Stern NJ, Bailey JS, and Cox NA. Incidence of Clostridium perfringens in

Broiler Chickens and Their Environment during Production and Processing.

Avian Dis. 45(4): 887-896, 2001b.

de Jong AE, Beumer RR, and Rombouts FM. Optimizing sporulation of Clostridium

perfringens. J Food Prot. 65(9):1457-1462, 2002.

Duncan CL, Strong DH. Improved medium for sporulation of Clostridium perfringens.

Appl Microbiol. 16(1):82-9. 1968.

Fenstermacher R. and Pomeroy BS. Clostridium infection in turkeys. Cornell Vet. 29: 25-

28. 1939.

Ficken. MD. Gangrenous dermatitis. In Diseases of Poultry. 9th Ed. B.W. Calneck., H.J.

Barnes, C. W. Beard., W. M. Reid and H.W. Yoder Jr., eds. Iowa State

University Press, Ames, IA. pp 268-270. 1991.

Frazier MN, Parizek W J., and Garner E. Gangrenous Dermatitis of Chickens. Avian

Dis. 8(2):269-273, 1964.

Gerdon. D. Effects of a mixed Clostridial Bacterin on Incidence of Gangrenous

dermatitis. Avian Dis. 17:205-206, 1973.

Gomis S, Amoako AK, Ngeleka AM, Belanger L, Althouse B, Kumor L, Waters E,

Stephens S, Riddell C, Potter A, and Allan B. Histopathologic and bacteriologic

116

http://www.jstor.org/action/doBasicSearch?Query=au:%22Stern%2C+N.+J.%22

evaluations of cellulitis detected in legs and caudal abdominal regions of

turkeys. Avian Dis. 46(1):192-197, 2002.

Gomis SM, Goodhope R, Kumor L, Caddy N, Riddell C, Potter AA, and Allan BJ.

Isolation of Escherichia coli from cellulitis and other lesions of the same bird in

broilers at slaughter. Can Vet J. 38(3):159-162, 1997.

Gubash SM. Synergistic haemolysis test for presumptive identification and differentiation

of Clostridium perfringens, C. bifermentans, C. sordellii, and C. paraperfringens.

J Clin Pathol. 33(4):395-399, 1980.

Heier B. T, A. Lovland, K. B. Soleim, M. Kaldhusal, and J. Jarp A Field Study of

Naturally Occurring Specific Antibodies against Clostridium perfringens Alpha

Toxin in Norwegian Broiler Flocks. Avian Dis, 45(3), 724-732, 2001.

Helfer DH, Dickinson EM, and Smith DH. Case report--Clostridium septicum infection

in a broiler flock. Avian Dis. 13(1):231-233, 1969.

Hofacre. C L, French JD, Page RK, and Fletcher OJ. Subcutaneous Clostridial infection

in Broilers. Avian Dis. 30:620-622, 1986.

Jeffrey JS, Nolan LK, Tonooka KH, Wolfe S, Giddings CW, Horne SM, Foley SL,

Lynne AM, Ebert JO, Elijah LM, Bjorklund G, Pfaff-McDonough SJ, Singer

RS, and Doetkott C. Virulence factors of Escherichia coli from cellulitis or

colisepticemia lesions in chickens. Avian Dis. 46(1): 48-52, 2002.

Johansson A, Aspan A, Bagge E, and Baverud V. Genetic diversity of Clostridium

perfringens type A isolates from animals, food poisoning outbreaks and sludge.

BMC Microbiol, 6: 47, 2006.

117

http://www.jstor.org/action/doBasicSearch?Query=au:%22Heier%2C+B.+T.%22

http://www.jstor.org/action/doBasicSearch?Query=au:%22Kaldhusal%2C+M.%22

http://www.jstor.org/action/doBasicSearch?Query=au:%22Jarp%2C+J.%22

Jost BH, Trinh HT, Songer JG. Clonal relationships among Clostridium perfringens of

porcine origin as determined by multilocus sequence typing. Vet Microbiol.

116(1-3):158-165, 2006.

Jost, BH, Billington SJ, Trinh HT, and Bueschel DM., Atypical cpb2 genes, encoding

beta2-toxin in Clostridium perfringens isolates of nonporcine origin. Infect

Immun 73: 652-656, 2005.

Kalender H., Isolation of Clostridium perfringens from chickens and detection of the

alpha toxin gene by Polymerase chain reaction (PCR). Turk. J vet. Anim Sci.

29:847-851, 2005.

Karasawa T, Wang X, Maegawa T, Michiwa Y, Kita H, Miwa K, and Nakamura S.

Clostridium sordellii phospholipase C: gene cloning and comparison of

enzymatic and biological activities with those of Clostridium perfringens and

Clostridium bifermentans phospholipase C. Infect Immun, 71(2):641-6, 2003.

Keyburn AL, Boyce JD., Vaz P., Bannam TL., Ford ME., Parker D., Rubbo AD., Rood

JI., and Moore RJ. NetB, a new toxin that is associated with avian necrotic

enteritis caused by Clostridium perfringens. PLoS pathogens 4(2) e26, 2008.

Keyburn AL, Sheedy SA, and Ford ME. Alpha-toxin of Clostridium perfringens is not an

essential virulence factor in necrotic enteritis in chickens. Infect Immun. 74(11):

6496-500, 2006.

Kotloff KL, Wasserman SS, Losonsky GA, et al. Safety and immunogenicity of

increasing doses of a Clostridium difficile toxoid vaccine administered to

healthy adults. Infect Immun. 69:988-995, 2001.

118

http://www.ncbi.nlm.nih.gov/pubmed/16650661


Kotloff KL, Wasserman SS, Losonsky GA, Thomas W Jr, Nichols R, Edelman R,

Bridwell M, and Monath TP. Safety and immunogenicity of increasing doses of

a Clostridium difficile toxoid vaccine administered to healthy adults. Infect

Immun. 69(2):988-95, 2001.

Kulkarni RR, Parreira VR, and Sharif S, et al. Clostridium perfringens Antigens

Recognized by Broiler Chickens Immune to Necrotic Enteritis. Clinical and

Vaccine Immunology, 13(12):1358–1362, 2006.

Kulkarni RR, Parreira VR, Sharif S, and Prescott JF. Immunization of broiler chickens

against Clostridium perfringens-induced necrotic enteritis. Clin Vaccine

Immunol. 14(9):1070-7, 2007.

Kumar MC. Prevention of turkey cellulitis. In: Proceedings of the 58th NCAD

Conference and symposium on lessons learned from AI preparation, pp.8. March

11-13, Saint Paul, MN, 2007.

Kumor LW, Olkowski AA, Gomis SM, and Allan BJ. Cellulitis in broiler chickens:

epidemiological trends, meat hygiene, and possible human health implications.

Avian Dis. 42(2):285-91, 1998.

Leclerc B, Fairbrother JM, Boulianne M, and Messier S. Evaluation of the adhesive

capacity of Escherichia coli isolates associated with avian cellulitis. Avian Dis.

47(1):21-31, 2003.

Li G., Hyun S. Lillehoj HS, Kyung Woo Lee, Seung IJ, Marc P, Gay, Ritter G. D.

Bautista D A, Phillips K, Neumann A P.6, Rehberger T G and Siragusa GR. An

outbreak of gangrenous dermatitis in commercial broiler chickens Avian

Pathology 39(4), 247-253, 2010a.

119

Li G., Hyun S. Lillehoj HS, Lee KW, Jang, SI, Marc P, Gay, Ritter G. D. Bautista D A,

Phillips K, Neumann A P., Rehberger T G and Siragusa GR. Immunopathology

and cytokine responses in commercial broiler chickens with gangrenous

dermatitis. Avian Pathology 39(4), 255-2564, 2010b.

Martel A, Devriese LA, Cauwerts K, De Gussem K, Decostere A, and Haesebrouck F.

Susceptibility of Clostridium perfringens strains from broiler chickens to

antibiotics and anticoccidials. Avian Pathol. 33 (1), 3-7, 2004.

McCourt MT, Finlay DA, Laird C, Smyth JA, Bell C, Ball HJ. Sandwich ELISA

detection of Clostridium perfringens cells and alpha-toxin from field cases of

necrotic enteritis of poultry. Vet Microbiol. 106(3-4):259-64. 2005.

McDevitt RM, Brooker JD, Acamovic T and Sparks NHC. Necrotic enteritis: a

continuing challenge for the poultry industry. Worlds Poultry science journal

62:221-247, 2006.

Meer RR and Songer JG. Multiplex polymerase chain reaction assay for genotyping

Clostridium perfringens enterotoxin. Rev Environ. Contam. Toxicol. 150:75-94,

1997.

Messier S, Quessy S, Robinson Y, Devriese LA, Hommez J, and Fairbrother JM. Focal

dermatitis and cellulitis in broiler chickens: bacteriological and pathological

findings. Avian Dis. 37(3):839-44, 1993.

Neeson BN., Graeme C. Clark, Helen S. Atkins, Bryan Lingard and Richard W. Titball.

Analysis of protection afforded by a Clostridium perfringens -toxoid against

heterologous clostridial phospholipases C. Microbial Pathogenesis 43(4):161-

165, 2007.

120

javascript:AL_get(this,%20'jour',%20'Avian%20Pathol.');

http://www.sciencedirect.com/science/journal/08824010

Neumann AP, Rehberger TG. MLST analysis reveals a highly conserved core genome

among poultry isolates of Clostridium septicum. Anaerobe. 15(3):99-106, 2009.

Ninomiya M, O Matsushita, J Minami, H Sakamoto, M Nakano, and A Okabe Role of

alpha-toxin in Clostridium perfringens infection determined by using

recombinants of C. perfringens and Bacillus subtilis. Infection and Immunity. 62

(11), 5032-5039, 1994.

Olkowski AA, Kumor L, Johnson D, Bielby M, Chirino-Trejo M, and Classen HL.

Cellulitis lesions in commercial turkeys identified during processing. Vet

Rec.145 (8):228-9, 1999.

Olkowski AA, Wojnarowicz C, Chirino-Trejo M, and Drew MD. Responses of broiler

chickens orally challenged with Clostridium perfringens isolated from field

cases of necrotic enteritis. Res Vet Sci. 81(1):99-108, 2006.

Omderdonk. A and Allen S.. Clostridium. In Manual of Clinical Microbiology 6th Ed. PR

Murray, E J Baron, M A Pfaller, F C TEnover and R H Yolken eds. American

Society for Microbiology, Washington DC. Pp 574-585, 1995.

Pedersen K, Bjerrum L, Heuer OE, Wong DM, and Nauerby B. Reproducible infection

model for Clostridium perfringens in broiler chickens. Avian Dis. 52(1):34-9,

2008.

Reed L.J and Muench H. A simple method of estimating fifty percent end-points. Am. J

Hyg. 27:493-497, 1938.

Rood, J. I. Virulence genes of Clostridium perfringens. Annu. Rev. Microbiol. 52:333–

360, 1998.

121



Rooney AP, Swezey JL, Friedman R, Hecht DW and Maddox CW. Analysis of

housekeeping virulence genes reveal cryptic lineages of Clostridium perfringens

that are associated with distinct disease presentations. Genetics 172:2081-2092,

2006.

Sasaki Y., K. Yamamoto, K. Amimoto, A. Kojima, Y. Ogikubo, M. Norimatsu, H. Ogata

and Y. Tamura, Amplification of the 16S-23S r DNA spacer region for rapid

detection of Clostridium chauvoei and Clostridium septicum. Research in

Veterinary Science, 71(3):227-229, 2001.

Saunders JR, and Bickford A. Clostridial infections of growing chickens. Avian Dis 9:

317-326, 1965.

Sawires YS, and Songer JG.Clostridium perfringens: insight into virulence evolution and

population structure. Anaerobe. 12 (1):23-43, 2006.

Shane S and van Der Sluis W. Global disease update 2002: more problems to be solved.

World Poultry18: 28-31, 2002.

Sheedy SA, and Ford ME. Alpha-toxin of Clostridium perfringens is not an essential

virulence factor in necrotic enteritis in chickens. Infect Immun. 74(11):6496-

500, 2006.

Shih NJ, and Labbé RG. Sporulation-promoting ability of Clostridium perfringens culture

fluids. Appl Environ Microbiol. 62(4):1441-3, 1996.

Shimizu, T, Ohtani K, Hirakawa H, Ohshima K, Yamashita A, Shiba T, Ogasawara N,

Hattori M, Kuhara S, and Hayashi H. Complete genome sequence of Clostridium

perfringens, an anaerobic flesh-eater. PNAS 99(2):996–1001, 2002.

122



http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%237137%232001%23999289996%23316453%23FLP%23&_cdi=7137&_pubType=J&_auth=y&_acct=C000032378&_version=1&_urlVersion=0&_userid=616288&md5=ceef87f5509851c525448d746b45a439

Smedley J. G.III, D. J. Fisher., S. Sayeed., and G. Chakrabarti. The enteric toxins of

Clostridium perfringens Rev Physiol Biochem Pharmacol, 152:183–204, 2004.

Smith-Slatas CL, Bourque M, and Salazar JC. Clostridium septicum infections in

children: a case report and review of the literature. Pediatrics. 117(4):796-805,

2006.

Songer JG. Clostridial enteric disease of domestic animals. Clin. Microbiol. Rev. 9, 216-

234, 1996.

Springer S, and Selbitz HJ. The control of necrotic enteritis in sucking piglets by means

of a Clostridium perfringens toxoid vaccine. FEMS Immunology & Medical

Microbiology, 24 (3):333-336, 1999.

Stevens DL, Titball RW, Jepson M, Bayer CR, Hayes-Schroer SM, and Bryant AE.

Immunization with the C-Domain of alpha -Toxin prevents lethal infection,

localizes tissue injury, and promotes host response to challenge with Clostridium

perfringens. J Infect Dis. 190(4):767-773, 2004.

St-Hilaire S, Arellano S, and Ribble C. Association between cellulitis (enlarged sternal

bursa) and focal ulcerative dermatitis in Ontario turkeys at the time of

processing. Avian Dis. 47(3):531-536, 2003.

Susan. E. Aiello. Breast blisters In: Mercks Veterinary Manual 8th ed. pp 675-676,

Merck and Co. Inc NJ, 2003.

Takeuchi S, Hashizume N, Kinoshita T, Kaidoh T, Tamura Y. Detection of Clostridium

septicum hemolysin gene by polymerase chain reaction. J Vet Med Sci,

59(9):853-855, 1997.

123

Tellez G, Pumford NR, Morgan MJ et al. Evidence for Clostridium septicum as a primary

cause of cellulitis in commercial turkeys. J. Vet. Diagn. Invest. 21:374-377,

2009.

Thachil AJ, McComb B, Andersen M, et al. Control of turkey Clostridial dermatitis using

a Clostridium septicum toxoid. In: Proc. of the 60th North Central Avian Disease

Conference, 11, Saint Paul, MN, USA, 2009.

Thompson DR, Parreira VR, Kulkarni RR, and Prescott JF. Live attenuated vaccine-

based control of necrotic enteritis of broiler chickens. Vet Microbiol. 113(1-

2):25-34, 2006

Timoney JF, Gillespie. JH., Scott F. and Barlough JE, In Hagan and Bruner’s

Microbiology and Infectious diseases of domestic Animals. 8th Ed. Comstock

publishing Associates, London. pp 214-240, 1998.

Titball RW, Naylor CE, and Basak AK. The Clostridium perfringens alpha-toxin.

Anaerobe. 5(2), 51-64, 1999.

Torres JF, Lyerly DM, Hill JE, and Monath TP. Evaluation of formalin-inactivated

Clostridium difficile vaccines administered by parenteral and mucosal routes of

immunization in hamsters. Infect Immun. 63(12):4619-4627, 1995.

Truscott RB, and Al-Sheikhly F. Reproduction and treatment of necrotic enteritis in

broilers. Am. J Vet Res. 38(6):857-861, 1977.

Van Immerseel F, DeBuck J, Pasmans F, et al. Clostridium perfringens in poultry: an

emerging threat for animal and public health. Avian pathol. 33:537-549, 2004.

124

javascript:AL_get(this,%20'jour',%20'Vet%20Microbiol.');

van-Immerseel FV, Buck JD., Pasmans F., Huyghebaert G, Haesebrouck F, and Richard

Ducatelle. Clostridium perfringens in poultry: an emerging threat for animal and

public health. Avian pathology, 33(6):537-549, 2004.

Varga J, Stirewalt VL, and Melville SB. The CcpA protein is necessary for efficient

sporulation and enterotoxin gene (cpe) regulation in Clostridium perfringens. J

Bacteriol. 186(16):5221-5229, 2004.

Wang RF, Cao WW, Franklin W, Campbell W, and Cerniglia CE. A 16S rDNA-based

PCR method for rapid and specific detection of Clostridium perfringens in food.

Mol Cell Probes. 8(2):131-7, 1994.

Wilder TD, Barbaree JM, Macklin KS, and Norton RA. Differences in the pathogenicity

of various bacterial isolates used in an induction model for gangrenous

dermatitis in broiler chickens. Avian Dis. 45(3):659-662, 2001.

Willoughby DH, Bickford AA, Cooper GL, and Charlton BR. Periodic recurrence of

gangrenous dermatitis associated with Clostridium septicum in a broiler chicken

operation. J Vet Diagn Invest. 8(2):259-261, 1996.

125

APPENDIX

AUTHORIZATION OF USE OF PUBLISHED MATERIAL

126

127

Documents

CELLULITIS IN TURKEYS: CHARACTERIZATION OF …